No_Show_Appointment.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Author: Zao Richardson Ansah* <br>\n",
    "*Project Date: 3/09/2022 - 6/09/2022*\n",
    "\n",
    "# Project: Appointments \"No Show\" \n",
    "\n",
    "## Table of Contents\n",
    "<ul>\n",
    "<li><a href=\"#intro\">Introduction</a></li>\n",
    "<li><a href=\"#wrangling\">Data Wrangling</a></li>\n",
    "<li><a href=\"#eda\">Exploratory Data Analysis</a></li>\n",
    "<li><a href=\"#conclusions\">Conclusions</a></li>\n",
    "</ul>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='intro'></a>\n",
    "## Introduction\n",
    "\n",
    "In this project we will be analyzing the [Appointments No-Show dataset](https://d17h27t6h515a5.cloudfront.net/topher/2017/October/59dd2e9a_noshowappointments-kagglev2-may-2016/noshowappointments-kagglev2-may-2016.csv), based on medical appointments in Brazil. In particularly, we will try to investigate if there are any factors that influence the appointment attendance (independent variable).\n",
    "\n",
    "Here are the research questions that will guide the analysis:\n",
    "\n",
    "1. Does gender impact the attendance?\n",
    "2. Is there a specific age group with higher attendance rate?\n",
    "3. How does the date difference between the scheduled and the appointment day impact the attendance rate? Does same day appointment have a higher attendance rate (which could be viewed as emergency visits)?\n",
    "4. Are there specific neighborhoods associated with higher No-Show?\n",
    "0. Does receiving a scholarship (Bolsa Familia) has any impact on the attendance?\n",
    "5. Does any medical condition impact the attendance rate?\n",
    "6. Does receiving the SMS reminder has any impact on the attendance?\n",
    "8. Is it possible to, based on the available features, predict whether a patient will show up to their scheduled appointment?\n",
    "\n",
    "> **Limitation**: it is important to highlight that, without the proper hypothesis testing techniques, the conclusions based solely on descriptive statistics are merely tentative."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='wrangling'></a>\n",
    "## Data Wrangling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>PatientId</th>\n",
       "      <th>AppointmentID</th>\n",
       "      <th>Gender</th>\n",
       "      <th>ScheduledDay</th>\n",
       "      <th>AppointmentDay</th>\n",
       "      <th>Age</th>\n",
       "      <th>Neighbourhood</th>\n",
       "      <th>Scholarship</th>\n",
       "      <th>Hipertension</th>\n",
       "      <th>Diabetes</th>\n",
       "      <th>Alcoholism</th>\n",
       "      <th>Handcap</th>\n",
       "      <th>SMS_received</th>\n",
       "      <th>No-show</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2.987250e+13</td>\n",
       "      <td>5642903</td>\n",
       "      <td>F</td>\n",
       "      <td>2016-04-29T18:38:08Z</td>\n",
       "      <td>2016-04-29T00:00:00Z</td>\n",
       "      <td>62</td>\n",
       "      <td>JARDIM DA PENHA</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5.589978e+14</td>\n",
       "      <td>5642503</td>\n",
       "      <td>M</td>\n",
       "      <td>2016-04-29T16:08:27Z</td>\n",
       "      <td>2016-04-29T00:00:00Z</td>\n",
       "      <td>56</td>\n",
       "      <td>JARDIM DA PENHA</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.262962e+12</td>\n",
       "      <td>5642549</td>\n",
       "      <td>F</td>\n",
       "      <td>2016-04-29T16:19:04Z</td>\n",
       "      <td>2016-04-29T00:00:00Z</td>\n",
       "      <td>62</td>\n",
       "      <td>MATA DA PRAIA</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      PatientId  AppointmentID Gender          ScheduledDay  \\\n",
       "0  2.987250e+13        5642903      F  2016-04-29T18:38:08Z   \n",
       "1  5.589978e+14        5642503      M  2016-04-29T16:08:27Z   \n",
       "2  4.262962e+12        5642549      F  2016-04-29T16:19:04Z   \n",
       "\n",
       "         AppointmentDay  Age    Neighbourhood  Scholarship  Hipertension  \\\n",
       "0  2016-04-29T00:00:00Z   62  JARDIM DA PENHA            0             1   \n",
       "1  2016-04-29T00:00:00Z   56  JARDIM DA PENHA            0             0   \n",
       "2  2016-04-29T00:00:00Z   62    MATA DA PRAIA            0             0   \n",
       "\n",
       "   Diabetes  Alcoholism  Handcap  SMS_received No-show  \n",
       "0         0           0        0             0      No  \n",
       "1         0           0        0             0      No  \n",
       "2         0           0        0             0      No  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(\"NoShowAppointment.csv\")\n",
    "df.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First, let's make sure all column names are lower case and separated by underscore:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['patientid', 'appointmentid', 'gender', 'scheduledday',\n",
       "       'appointmentday', 'age', 'neighbourhood', 'scholarship', 'hipertension',\n",
       "       'diabetes', 'alcoholism', 'handcap', 'sms_received', 'no-show'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns.str.lower()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Manually editing the col names for readibility and typos\n",
    "df.columns = ['patient_id', 'appointment_id', 'gender', 'scheduled_day',\n",
    "       'appointment_day', 'age', 'neighborhood', 'scholarship', 'hypertension',\n",
    "       'diabetes', 'alcoholism', 'handicap', 'sms_received', 'no_show']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(110527, 14)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Checking shape of dataset : rows vs cols\n",
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This dataset contains 110,527 rows of medical appointment data, with 14 Features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 110527 entries, 0 to 110526\n",
      "Data columns (total 14 columns):\n",
      " #   Column           Non-Null Count   Dtype  \n",
      "---  ------           --------------   -----  \n",
      " 0   patient_id       110527 non-null  float64\n",
      " 1   appointment_id   110527 non-null  int64  \n",
      " 2   gender           110527 non-null  object \n",
      " 3   scheduled_day    110527 non-null  object \n",
      " 4   appointment_day  110527 non-null  object \n",
      " 5   age              110527 non-null  int64  \n",
      " 6   neighborhood     110527 non-null  object \n",
      " 7   scholarship      110527 non-null  int64  \n",
      " 8   hypertension     110527 non-null  int64  \n",
      " 9   diabetes         110527 non-null  int64  \n",
      " 10  alcoholism       110527 non-null  int64  \n",
      " 11  handicap         110527 non-null  int64  \n",
      " 12  sms_received     110527 non-null  int64  \n",
      " 13  no_show          110527 non-null  object \n",
      "dtypes: float64(1), int64(8), object(5)\n",
      "memory usage: 11.8+ MB\n"
     ]
    }
   ],
   "source": [
    "# Checking missing data and data types\n",
    "df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There is no missing data - which is a good thing!\n",
    "\n",
    "We need to fix the data type of some columns:\n",
    "\n",
    "+ **Gender**: String -> Category (M: Male | F: Female)\n",
    "+ **Scheduled Day** / **Appointment Day**: String -> Date\n",
    "+ **No-Show**: String -> Integer (1: Yes | 0: No)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0, 48228, 0)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Checking for duplicates\n",
    "df.duplicated().sum() , df.duplicated('patient_id').sum(), df.duplicated('appointment_id').sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are no duplicated rows. Although there are 48,228 patients that scheduled more than one appointment. All Appointment IDs are unique values.\n",
    "\n",
    "The columns \"Patient ID\" and \"Appointment ID\" doesn't seem very promising to our analysis, so we can remove them as well:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.drop(['patient_id','appointment_id'],axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>scholarship</th>\n",
       "      <th>hypertension</th>\n",
       "      <th>diabetes</th>\n",
       "      <th>alcoholism</th>\n",
       "      <th>handicap</th>\n",
       "      <th>sms_received</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>110527.000000</td>\n",
       "      <td>110527.000000</td>\n",
       "      <td>110527.000000</td>\n",
       "      <td>110527.000000</td>\n",
       "      <td>110527.000000</td>\n",
       "      <td>110527.000000</td>\n",
       "      <td>110527.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>37.088874</td>\n",
       "      <td>0.098266</td>\n",
       "      <td>0.197246</td>\n",
       "      <td>0.071865</td>\n",
       "      <td>0.030400</td>\n",
       "      <td>0.022248</td>\n",
       "      <td>0.321026</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>23.110205</td>\n",
       "      <td>0.297675</td>\n",
       "      <td>0.397921</td>\n",
       "      <td>0.258265</td>\n",
       "      <td>0.171686</td>\n",
       "      <td>0.161543</td>\n",
       "      <td>0.466873</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>18.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>37.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>55.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>115.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 age    scholarship   hypertension       diabetes  \\\n",
       "count  110527.000000  110527.000000  110527.000000  110527.000000   \n",
       "mean       37.088874       0.098266       0.197246       0.071865   \n",
       "std        23.110205       0.297675       0.397921       0.258265   \n",
       "min        -1.000000       0.000000       0.000000       0.000000   \n",
       "25%        18.000000       0.000000       0.000000       0.000000   \n",
       "50%        37.000000       0.000000       0.000000       0.000000   \n",
       "75%        55.000000       0.000000       0.000000       0.000000   \n",
       "max       115.000000       1.000000       1.000000       1.000000   \n",
       "\n",
       "          alcoholism       handicap   sms_received  \n",
       "count  110527.000000  110527.000000  110527.000000  \n",
       "mean        0.030400       0.022248       0.321026  \n",
       "std         0.171686       0.161543       0.466873  \n",
       "min         0.000000       0.000000       0.000000  \n",
       "25%         0.000000       0.000000       0.000000  \n",
       "50%         0.000000       0.000000       0.000000  \n",
       "75%         0.000000       0.000000       1.000000  \n",
       "max         1.000000       4.000000       1.000000  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Looking at the overall description of the data, the column Age seems off: the minimum value is -1, which doesn't make sense. In addition, the maximum value is 115, which seems too high. We need to investigate for possible outlier and remove negative values:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>scheduled_day</th>\n",
       "      <th>appointment_day</th>\n",
       "      <th>age</th>\n",
       "      <th>neighborhood</th>\n",
       "      <th>scholarship</th>\n",
       "      <th>hypertension</th>\n",
       "      <th>diabetes</th>\n",
       "      <th>alcoholism</th>\n",
       "      <th>handicap</th>\n",
       "      <th>sms_received</th>\n",
       "      <th>no_show</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>99832</th>\n",
       "      <td>F</td>\n",
       "      <td>2016-06-06T08:58:13Z</td>\n",
       "      <td>2016-06-06T00:00:00Z</td>\n",
       "      <td>-1</td>\n",
       "      <td>ROMÃO</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      gender         scheduled_day       appointment_day  age neighborhood  \\\n",
       "99832      F  2016-06-06T08:58:13Z  2016-06-06T00:00:00Z   -1        ROMÃO   \n",
       "\n",
       "       scholarship  hypertension  diabetes  alcoholism  handicap  \\\n",
       "99832            0             0         0           0         0   \n",
       "\n",
       "       sms_received no_show  \n",
       "99832             0      No  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df.age < 0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Outlier Threshold: 110\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>scheduled_day</th>\n",
       "      <th>appointment_day</th>\n",
       "      <th>age</th>\n",
       "      <th>neighborhood</th>\n",
       "      <th>scholarship</th>\n",
       "      <th>hypertension</th>\n",
       "      <th>diabetes</th>\n",
       "      <th>alcoholism</th>\n",
       "      <th>handicap</th>\n",
       "      <th>sms_received</th>\n",
       "      <th>no_show</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>63912</th>\n",
       "      <td>F</td>\n",
       "      <td>2016-05-16T09:17:44Z</td>\n",
       "      <td>2016-05-19T00:00:00Z</td>\n",
       "      <td>115</td>\n",
       "      <td>ANDORINHAS</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>Yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>63915</th>\n",
       "      <td>F</td>\n",
       "      <td>2016-05-16T09:17:44Z</td>\n",
       "      <td>2016-05-19T00:00:00Z</td>\n",
       "      <td>115</td>\n",
       "      <td>ANDORINHAS</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>Yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>68127</th>\n",
       "      <td>F</td>\n",
       "      <td>2016-04-08T14:29:17Z</td>\n",
       "      <td>2016-05-16T00:00:00Z</td>\n",
       "      <td>115</td>\n",
       "      <td>ANDORINHAS</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>Yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>76284</th>\n",
       "      <td>F</td>\n",
       "      <td>2016-05-30T09:44:51Z</td>\n",
       "      <td>2016-05-30T00:00:00Z</td>\n",
       "      <td>115</td>\n",
       "      <td>ANDORINHAS</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97666</th>\n",
       "      <td>F</td>\n",
       "      <td>2016-05-19T07:57:56Z</td>\n",
       "      <td>2016-06-03T00:00:00Z</td>\n",
       "      <td>115</td>\n",
       "      <td>SÃO JOSÉ</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      gender         scheduled_day       appointment_day  age neighborhood  \\\n",
       "63912      F  2016-05-16T09:17:44Z  2016-05-19T00:00:00Z  115   ANDORINHAS   \n",
       "63915      F  2016-05-16T09:17:44Z  2016-05-19T00:00:00Z  115   ANDORINHAS   \n",
       "68127      F  2016-04-08T14:29:17Z  2016-05-16T00:00:00Z  115   ANDORINHAS   \n",
       "76284      F  2016-05-30T09:44:51Z  2016-05-30T00:00:00Z  115   ANDORINHAS   \n",
       "97666      F  2016-05-19T07:57:56Z  2016-06-03T00:00:00Z  115     SÃO JOSÉ   \n",
       "\n",
       "       scholarship  hypertension  diabetes  alcoholism  handicap  \\\n",
       "63912            0             0         0           0         1   \n",
       "63915            0             0         0           0         1   \n",
       "68127            0             0         0           0         1   \n",
       "76284            0             0         0           0         1   \n",
       "97666            0             1         0           0         0   \n",
       "\n",
       "       sms_received no_show  \n",
       "63912             0     Yes  \n",
       "63915             0     Yes  \n",
       "68127             0     Yes  \n",
       "76284             0      No  \n",
       "97666             1      No  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Box-plot to visualize threshold for possible outliers\n",
    "df.age.plot(kind='box');\n",
    "iqr = df.age.quantile(0.75) - df.age.quantile(0.25)\n",
    "outlier = df.age.quantile(0.75) + 1.5*iqr\n",
    "print('Outlier Threshold:', int(round(outlier,0)))\n",
    "df[df.age > 110]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Looks like there is only one negative entry. And five entries with age greater than the ourlier threshold of 110 years.\n",
    "Let's go ahead and remove them:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(110521, 12)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = df[(df.age >=0) & (df.age <=110)]\n",
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's tweak this dataset a little more by renaming the column \"no_show\" as \"attended\", since it is more natural to think of 'yes' as a positive observation.\n",
    "\n",
    "We will rename the column and replace the observations No/Yes with 1/0:\n",
    "\n",
    "> *Obs: since \"attended\" is going to be our dependent variable, storing it as a binary 1/0 will make the analysis easier, because the mean will automatically represent the proportion.*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>scheduled_day</th>\n",
       "      <th>appointment_day</th>\n",
       "      <th>age</th>\n",
       "      <th>neighborhood</th>\n",
       "      <th>scholarship</th>\n",
       "      <th>hypertension</th>\n",
       "      <th>diabetes</th>\n",
       "      <th>alcoholism</th>\n",
       "      <th>handicap</th>\n",
       "      <th>sms_received</th>\n",
       "      <th>attended</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>F</td>\n",
       "      <td>2016-04-29T18:38:08Z</td>\n",
       "      <td>2016-04-29T00:00:00Z</td>\n",
       "      <td>62</td>\n",
       "      <td>JARDIM DA PENHA</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  gender         scheduled_day       appointment_day  age     neighborhood  \\\n",
       "0      F  2016-04-29T18:38:08Z  2016-04-29T00:00:00Z   62  JARDIM DA PENHA   \n",
       "\n",
       "   scholarship  hypertension  diabetes  alcoholism  handicap  sms_received  \\\n",
       "0            0             1         0           0         0             0   \n",
       "\n",
       "   attended  \n",
       "0         1  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.rename(columns={'no_show': 'attended'}, inplace=True)\n",
    "df.attended.replace(['No', 'Yes'], [1, 0], inplace=True)\n",
    "df.head(1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Good. Our changes look good so far.\n",
    "\n",
    "Now, let's fix the data types:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.gender = df.gender.astype('category')\n",
    "df.scheduled_day = pd.to_datetime(df.scheduled_day)\n",
    "df.appointment_day = pd.to_datetime(df.appointment_day)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 110521 entries, 0 to 110526\n",
      "Data columns (total 12 columns):\n",
      " #   Column           Non-Null Count   Dtype              \n",
      "---  ------           --------------   -----              \n",
      " 0   gender           110521 non-null  category           \n",
      " 1   scheduled_day    110521 non-null  datetime64[ns, UTC]\n",
      " 2   appointment_day  110521 non-null  datetime64[ns, UTC]\n",
      " 3   age              110521 non-null  int64              \n",
      " 4   neighborhood     110521 non-null  object             \n",
      " 5   scholarship      110521 non-null  int64              \n",
      " 6   hypertension     110521 non-null  int64              \n",
      " 7   diabetes         110521 non-null  int64              \n",
      " 8   alcoholism       110521 non-null  int64              \n",
      " 9   handicap         110521 non-null  int64              \n",
      " 10  sms_received     110521 non-null  int64              \n",
      " 11  attended         110521 non-null  int64              \n",
      "dtypes: category(1), datetime64[ns, UTC](2), int64(8), object(1)\n",
      "memory usage: 10.2+ MB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There is one last check before moving on to analyze the data: what do the datetime columns look  like?\n",
    "\n",
    "Let's create a new column calculating the difference between the date of the appointment and when it was scheduled:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count                        110521\n",
       "mean      9 days 17:08:35.550203128\n",
       "std      15 days 05:51:35.542808416\n",
       "min               -7 days +10:10:40\n",
       "25%               -1 days +15:41:32\n",
       "50%                 3 days 11:22:48\n",
       "75%                14 days 07:41:28\n",
       "max               178 days 13:19:01\n",
       "Name: delta_date, dtype: object"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['delta_date'] = df.appointment_day - df.scheduled_day\n",
    "df.delta_date.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Strange... looks like there are some negative values. We need to further investigate this:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>scheduled_day</th>\n",
       "      <th>appointment_day</th>\n",
       "      <th>delta_date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2016-04-29 18:38:08+00:00</td>\n",
       "      <td>2016-04-29 00:00:00+00:00</td>\n",
       "      <td>-1 days +05:21:52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2016-04-29 16:08:27+00:00</td>\n",
       "      <td>2016-04-29 00:00:00+00:00</td>\n",
       "      <td>-1 days +07:51:33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2016-04-29 16:19:04+00:00</td>\n",
       "      <td>2016-04-29 00:00:00+00:00</td>\n",
       "      <td>-1 days +07:40:56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2016-04-29 17:29:31+00:00</td>\n",
       "      <td>2016-04-29 00:00:00+00:00</td>\n",
       "      <td>-1 days +06:30:29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2016-04-29 16:07:23+00:00</td>\n",
       "      <td>2016-04-29 00:00:00+00:00</td>\n",
       "      <td>-1 days +07:52:37</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              scheduled_day           appointment_day        delta_date\n",
       "0 2016-04-29 18:38:08+00:00 2016-04-29 00:00:00+00:00 -1 days +05:21:52\n",
       "1 2016-04-29 16:08:27+00:00 2016-04-29 00:00:00+00:00 -1 days +07:51:33\n",
       "2 2016-04-29 16:19:04+00:00 2016-04-29 00:00:00+00:00 -1 days +07:40:56\n",
       "3 2016-04-29 17:29:31+00:00 2016-04-29 00:00:00+00:00 -1 days +06:30:29\n",
       "4 2016-04-29 16:07:23+00:00 2016-04-29 00:00:00+00:00 -1 days +07:52:37"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Looking only at negative values\n",
    "df[df.delta_date < pd.Timedelta(0, 'D')][['scheduled_day','appointment_day', 'delta_date']].head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ok... the -1 day difference can be explained due to the fact that Appointment Day column does not contain information regarding the time of the appointment.\n",
    "\n",
    "For this particular case, we can assume the appointment was actually scheduled on the same day. Let's fix the \"Delta Date\" column to reflect this assumption:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "min      -1 days +03:15:06\n",
       "max      -1 days +17:50:24\n",
       "count                38561\n",
       "Name: delta_date, dtype: object"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mask = ((df.delta_date <= pd.Timedelta(0, 'D')) & (df.delta_date >= pd.Timedelta(-1, 'D')))\n",
    "df[mask].delta_date.aggregate([min,max, 'count']) # 38562 occurrences of value: -1 day"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>scheduled_day</th>\n",
       "      <th>appointment_day</th>\n",
       "      <th>age</th>\n",
       "      <th>neighborhood</th>\n",
       "      <th>scholarship</th>\n",
       "      <th>hypertension</th>\n",
       "      <th>diabetes</th>\n",
       "      <th>alcoholism</th>\n",
       "      <th>handicap</th>\n",
       "      <th>sms_received</th>\n",
       "      <th>attended</th>\n",
       "      <th>delta_date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>F</td>\n",
       "      <td>2016-04-29 18:38:08+00:00</td>\n",
       "      <td>2016-04-29 00:00:00+00:00</td>\n",
       "      <td>62</td>\n",
       "      <td>JARDIM DA PENHA</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0 days</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>M</td>\n",
       "      <td>2016-04-29 16:08:27+00:00</td>\n",
       "      <td>2016-04-29 00:00:00+00:00</td>\n",
       "      <td>56</td>\n",
       "      <td>JARDIM DA PENHA</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0 days</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>F</td>\n",
       "      <td>2016-04-29 16:19:04+00:00</td>\n",
       "      <td>2016-04-29 00:00:00+00:00</td>\n",
       "      <td>62</td>\n",
       "      <td>MATA DA PRAIA</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0 days</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>F</td>\n",
       "      <td>2016-04-29 17:29:31+00:00</td>\n",
       "      <td>2016-04-29 00:00:00+00:00</td>\n",
       "      <td>8</td>\n",
       "      <td>PONTAL DE CAMBURI</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0 days</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>F</td>\n",
       "      <td>2016-04-29 16:07:23+00:00</td>\n",
       "      <td>2016-04-29 00:00:00+00:00</td>\n",
       "      <td>56</td>\n",
       "      <td>JARDIM DA PENHA</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0 days</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  gender             scheduled_day           appointment_day  age  \\\n",
       "0      F 2016-04-29 18:38:08+00:00 2016-04-29 00:00:00+00:00   62   \n",
       "1      M 2016-04-29 16:08:27+00:00 2016-04-29 00:00:00+00:00   56   \n",
       "2      F 2016-04-29 16:19:04+00:00 2016-04-29 00:00:00+00:00   62   \n",
       "3      F 2016-04-29 17:29:31+00:00 2016-04-29 00:00:00+00:00    8   \n",
       "4      F 2016-04-29 16:07:23+00:00 2016-04-29 00:00:00+00:00   56   \n",
       "\n",
       "        neighborhood  scholarship  hypertension  diabetes  alcoholism  \\\n",
       "0    JARDIM DA PENHA            0             1         0           0   \n",
       "1    JARDIM DA PENHA            0             0         0           0   \n",
       "2      MATA DA PRAIA            0             0         0           0   \n",
       "3  PONTAL DE CAMBURI            0             0         0           0   \n",
       "4    JARDIM DA PENHA            0             1         1           0   \n",
       "\n",
       "   handicap  sms_received  attended delta_date  \n",
       "0         0             0         1     0 days  \n",
       "1         0             0         1     0 days  \n",
       "2         0             0         1     0 days  \n",
       "3         0             0         1     0 days  \n",
       "4         0             0         1     0 days  "
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Replacing values based on condition\n",
    "df.loc[mask, 'delta_date'] = pd.Timedelta(0, 'D')\n",
    "df[mask].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "min      0 days 00:00:00\n",
       "max      0 days 00:00:00\n",
       "count              38561\n",
       "Name: delta_date, dtype: object"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Checking for correct replacement\n",
    "df[mask].delta_date.aggregate([min,max, 'count'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The alteration on column Delta Date was successful: we substituted 38,562 entries from -1 days to 0 day - representing same day appointment.\n",
    "\n",
    "But we still need to check the other negative values:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>scheduled_day</th>\n",
       "      <th>appointment_day</th>\n",
       "      <th>age</th>\n",
       "      <th>neighborhood</th>\n",
       "      <th>scholarship</th>\n",
       "      <th>hypertension</th>\n",
       "      <th>diabetes</th>\n",
       "      <th>alcoholism</th>\n",
       "      <th>handicap</th>\n",
       "      <th>sms_received</th>\n",
       "      <th>attended</th>\n",
       "      <th>delta_date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>27033</th>\n",
       "      <td>M</td>\n",
       "      <td>2016-05-10 10:51:53+00:00</td>\n",
       "      <td>2016-05-09 00:00:00+00:00</td>\n",
       "      <td>38</td>\n",
       "      <td>RESISTÊNCIA</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-2 days +13:08:07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55226</th>\n",
       "      <td>F</td>\n",
       "      <td>2016-05-18 14:50:41+00:00</td>\n",
       "      <td>2016-05-17 00:00:00+00:00</td>\n",
       "      <td>19</td>\n",
       "      <td>SANTO ANTÔNIO</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-2 days +09:09:19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>64175</th>\n",
       "      <td>F</td>\n",
       "      <td>2016-05-05 13:43:58+00:00</td>\n",
       "      <td>2016-05-04 00:00:00+00:00</td>\n",
       "      <td>22</td>\n",
       "      <td>CONSOLAÇÃO</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-2 days +10:16:02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>71533</th>\n",
       "      <td>F</td>\n",
       "      <td>2016-05-11 13:49:20+00:00</td>\n",
       "      <td>2016-05-05 00:00:00+00:00</td>\n",
       "      <td>81</td>\n",
       "      <td>SANTO ANTÔNIO</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-7 days +10:10:40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>72362</th>\n",
       "      <td>M</td>\n",
       "      <td>2016-05-04 06:50:57+00:00</td>\n",
       "      <td>2016-05-03 00:00:00+00:00</td>\n",
       "      <td>7</td>\n",
       "      <td>TABUAZEIRO</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-2 days +17:09:03</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      gender             scheduled_day           appointment_day  age  \\\n",
       "27033      M 2016-05-10 10:51:53+00:00 2016-05-09 00:00:00+00:00   38   \n",
       "55226      F 2016-05-18 14:50:41+00:00 2016-05-17 00:00:00+00:00   19   \n",
       "64175      F 2016-05-05 13:43:58+00:00 2016-05-04 00:00:00+00:00   22   \n",
       "71533      F 2016-05-11 13:49:20+00:00 2016-05-05 00:00:00+00:00   81   \n",
       "72362      M 2016-05-04 06:50:57+00:00 2016-05-03 00:00:00+00:00    7   \n",
       "\n",
       "        neighborhood  scholarship  hypertension  diabetes  alcoholism  \\\n",
       "27033    RESISTÊNCIA            0             0         0           0   \n",
       "55226  SANTO ANTÔNIO            0             0         0           0   \n",
       "64175     CONSOLAÇÃO            0             0         0           0   \n",
       "71533  SANTO ANTÔNIO            0             0         0           0   \n",
       "72362     TABUAZEIRO            0             0         0           0   \n",
       "\n",
       "       handicap  sms_received  attended        delta_date  \n",
       "27033         1             0         0 -2 days +13:08:07  \n",
       "55226         1             0         0 -2 days +09:09:19  \n",
       "64175         0             0         0 -2 days +10:16:02  \n",
       "71533         0             0         0 -7 days +10:10:40  \n",
       "72362         0             0         0 -2 days +17:09:03  "
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df.delta_date < pd.Timedelta(0,'D')]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are 4 entries with negative two days and 1 entry with negative seven days. Since we can't come up with a reasonable explanation for these cases, we will remove them from the dataset:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(110521, 13)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(110516, 13)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = df[df.delta_date >= pd.Timedelta(0,'D')]\n",
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We completed the data cleaning process and transformation, ending up with 110,521 observations and 13 columns.\n",
    "\n",
    "Now we are finally ready to start investigating and exploring our dataset!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='eda'></a>\n",
    "## Exploratory Data Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Dataset Description"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "scholarship     0.10\n",
       "hypertension    0.20\n",
       "diabetes        0.07\n",
       "alcoholism      0.03\n",
       "handicap        0.02\n",
       "sms_received    0.32\n",
       "attended        0.80\n",
       "dtype: float64"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Quickly obtaining proportions for the binary features\n",
    "ref = ['scholarship', 'hypertension','diabetes', 'alcoholism', 'handicap',\n",
    "       'sms_received', 'attended']\n",
    "df[ref].mean().round(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "gender\n",
       "F    0.65\n",
       "M    0.35\n",
       "dtype: float64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(df.groupby('gender').size() / df.shape[0]).round(2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Looking at the proportion of the binary features we can see that, for this sample:\n",
    "+ There is almost 2 times more female than male patients;\n",
    "+ Around 10% of patients receive the Scholarship ('Bolsa Familia');\n",
    "+ About 20% of patients suffer from hypertension;\n",
    "+ 7% of patients have diabetes;\n",
    "+ Only 3% suffer from alcoholism;\n",
    "+ Around 2% are handicapped;\n",
    "+ Roughly a third of patients have received the SMS reminder;\n",
    "* Almost 80% of all patients did show up to their scheduled appointment."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZIAAAEICAYAAAB1f3LfAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAAbdklEQVR4nO3de7xdZX3n8c/XRBFQBOGAIRcPl3gBBoOJiPVSlCpRVGIHNIw1qaJRBl/q6LQNtFNoZzKFTpWKLVgQ5KLc5CKpkNYIDtR5cfFwqQSQeoBIDolJFISgGEz8zh/r2bpzcs5hJ+vss9nnfN+v13qdtX5rPWs9D+j+8Vz22rJNRETE9npepysQERHdLYkkIiJqSSKJiIhakkgiIqKWJJKIiKgliSQiImpJIokYQ5KWSVrY6XpEjKYkkpgQJP1fSY9L2qHNz/jooNjhkgYax7bfafvCFu5lSfu3o54Roy2JJMY9Sb3AmwED7+1sbTpP0uRO1yHGlySSmAgWALcCFwBbDCtJ2l3SP0t6UtL3Jf0vSd9rOv8qScslPSbpAUnvr1OR5l6LpP0l3STpCUk/lXR5id9cLv93SU9J+kCJf0xSf6nLUkl7N933HaV+T0g6q9y38Zw/lvT/JJ0h6THgVEn7SbpR0s/Ks78uadem+62U9CeSfiDpF5LOk7RXGZrbIOk7knar888ixo8kkpgIFgBfL9uRkvZqOvePwC+Al1Elmd8mGkk7A8uBS4A9geOAsyQdOEr1+p/At4HdgGnAlwBsv6Wcf43tF9m+XNLbgL8B3g9MAX4MXFbquQdwJXASsDvwAPB7g571euCh0o4lgMr99gZeDUwHTh1U5j8DbwdeAbwHWAacDOxB9dnxqZrtj3EiiSTGNUlvAl4OXGH7DuBB4L+Uc5OoPixPsf1L2/cBzfMX7wZW2v6q7U227wSuAo4Z4ZFnSvp5YwO+NcK1vy5129v2r2x/b4RrPwicb/tO2xupksYbyrDdu4B7bV9texNwJvCTQeVX2/5SacfTtvttL7e90fZ64AvA7w8q8yXba20/CvwbcJvtu8rzrwEOGaG+MYEkkcR4txD4tu2fluNL+F2voweYDKxqur55/+XA6wclhg9S9V6G8ynbuzY2qmQ0nD+l6hncLuleSR8Z4dq9qXohANh+CvgZMLWcW9V0zsDAoPLN7ULSnpIuk/SopCeBr1H1NJqtbdp/eojjF41Q35hAMukW45akHamGgiZJavwX+g7ArpJeA6wANlENK/1HOT+96RargJtsv70d9bP9E+Bjpa5vAr4j6Wbb/UNcvpoqsVGu35lqGOtRYA1VGxrn1HzceNyg478psYNt/0zSPOAfajUoJqz0SGI8mwdsBg4AZpXt1VTDNAtsbwauppp83knSq6jmUxq+BbxC0ockPb9sr5P06tGonKRjJTU+8B+n+mDfXI7XAvs2XX4J8GFJs8oS5v9NNdS0ErgO+E+S5pUVWScycq8J4MXAU8DPJU0F/mQ02hQTUxJJjGcLga/afsT2Txob1X95f7B86H4SeAnVnMLFwKXARgDbG4B3APOpegQ/AU6n6tWMhtcBt0l6ClgKfNr2w+XcqcCFZUjt/bZvAP4H1RzNGmC/Ui/KsN2xwN9SDXcdAPQ12jGMvwJeCzxBlYiuHqU2xQSk/LBVxO9IOh14me2u/fa5pOdRzZF80PZ3O12fGP/SI4kJrXxP5GBVDgWOp1qR1FUkHSlp1zLsdTLVJP6tHa5WTBCZbI+J7sVUw1l7A+uAzwPXdrRG2+cNVPMoLwDuA+bZfrqzVYqJIkNbERFRS4a2IiKilgk3tLXHHnu4t7e309WIiOgqd9xxx09t9wx1bsIlkt7eXvr6+jpdjYiIriLpx8Ody9BWRETUkkQSERG1JJFEREQtSSQREVFLEklERNSSRBIREbUkkURERC1JJBERUUsSSURE1DLhvtleR+/i6zr27JWnHdWxZ0dEjCQ9koiIqCWJJCIiakkiiYiIWpJIIiKiliSSiIioJYkkIiJqSSKJiIhakkgiIqKWJJKIiKilbYlE0vmS1kla0RS7XNLdZVsp6e4S75X0dNO5LzeVmS3pHkn9ks6UpBLfodyvX9Jtknrb1ZaIiBheO3skFwBzmwO2P2B7lu1ZwFXA1U2nH2ycs/2JpvjZwCJgZtka9zweeNz2/sAZwOltaUVERIyobYnE9s3AY0OdK72K9wOXjnQPSVOAXWzfYtvARcC8cvpo4MKyfyVwRKO3EhERY6dTcyRvBtba/lFTbB9Jd0m6SdKbS2wqMNB0zUCJNc6tArC9CXgC2H2oh0laJKlPUt/69etHsx0RERNepxLJcWzZG1kDzLB9CPBZ4BJJuwBD9TBc/o50bsugfY7tObbn9PT01Kh2REQMNuavkZc0GfhDYHYjZnsjsLHs3yHpQeAVVD2QaU3FpwGry/4AMB0YKPd8CcMMpUVERPt0okfyB8APbf92yEpSj6RJZX9fqkn1h2yvATZIOqzMfywAri3FlgILy/4xwI1lHiUiIsZQO5f/XgrcArxS0oCk48up+Ww9yf4W4AeS/p1q4vwTthu9ixOArwD9wIPAshI/D9hdUj/VcNjidrUlIiKG17ahLdvHDRP/4yFiV1EtBx7q+j7goCHivwKOrVfLiIioKz+1G/EckZ9yjm6VV6REREQtSSQREVFLEklERNSSRBIREbUkkURERC1JJBERUUsSSURE1JJEEhERtSSRRERELUkkERFRSxJJRETUkkQSERG1JJFEREQtSSQREVFLEklERNSSRBIREbUkkURERC1JJBERUUvbEomk8yWtk7SiKXaqpEcl3V22dzWdO0lSv6QHJB3ZFJ8t6Z5y7kxJKvEdJF1e4rdJ6m1XWyIiYnjt7JFcAMwdIn6G7Vllux5A0gHAfODAUuYsSZPK9WcDi4CZZWvc83jgcdv7A2cAp7erIRERMby2JRLbNwOPtXj50cBltjfafhjoBw6VNAXYxfYttg1cBMxrKnNh2b8SOKLRW4mIiLEzuQPP/KSkBUAf8DnbjwNTgVubrhkosV+X/cFxyt9VALY3SXoC2B346eAHSlpE1athxowZo9qYaJ/exdd15LkrTzuqI8+N6FZjPdl+NrAfMAtYA3y+xIfqSXiE+Ehltg7a59ieY3tOT0/PNlU4IiJGNqaJxPZa25tt/wY4Fzi0nBoApjddOg1YXeLThohvUUbSZOAltD6UFhERo2RME0mZ82h4H9BY0bUUmF9WYu1DNal+u+01wAZJh5X5jwXAtU1lFpb9Y4AbyzxKRESMobbNkUi6FDgc2EPSAHAKcLikWVRDUCuBjwPYvlfSFcB9wCbgRNuby61OoFoBtiOwrGwA5wEXS+qn6onMb1dbIiJieG1LJLaPGyJ83gjXLwGWDBHvAw4aIv4r4Ng6dYyIiPryzfaIiKilE8t/I57TOrXsOKJbpUcSERG1JJFEREQtSSQREVFLEklERNSSRBIREbUkkURERC1Z/hsjylLYiHg26ZFEREQtSSQREVFLEklERNSSRBIREbUkkURERC1JJBERUUuW/3aJLMONiOeq9EgiIqKWJJKIiKgliSQiImppWyKRdL6kdZJWNMX+j6QfSvqBpGsk7VrivZKelnR32b7cVGa2pHsk9Us6U5JKfAdJl5f4bZJ629WWiIgYXjt7JBcAcwfFlgMH2T4Y+A/gpKZzD9qeVbZPNMXPBhYBM8vWuOfxwOO29wfOAE4f/SZERMSzaVsisX0z8Nig2LdtbyqHtwLTRrqHpCnALrZvsW3gImBeOX00cGHZvxI4otFbiYiIsdPJOZKPAMuajveRdJekmyS9ucSmAgNN1wyUWOPcKoCSnJ4Adh/qQZIWSeqT1Ld+/frRbENExITXkUQi6c+BTcDXS2gNMMP2IcBngUsk7QIM1cNw4zYjnNsyaJ9je47tOT09PfUqHxERWxjzLyRKWgi8GziiDFdheyOwsezfIelB4BVUPZDm4a9pwOqyPwBMBwYkTQZewqChtIiIaL8x7ZFImgv8GfBe279sivdImlT296WaVH/I9hpgg6TDyvzHAuDaUmwpsLDsHwPc2EhMERExdtrWI5F0KXA4sIekAeAUqlVaOwDLy7z4rWWF1luAv5a0CdgMfMJ2o3dxAtUKsB2p5lQa8yrnARdL6qfqicxvV1siImJ4bUskto8bInzeMNdeBVw1zLk+4KAh4r8Cjq1Tx4iIqK+loS1JW32QR0REQOtzJF+WdLuk/9r4NnpERAS0mEhsvwn4INUqqT5Jl0h6e1trFhERXaHlVVu2fwT8BdWqq98HzizvzfrDdlUuIiKe+1qdIzlY0hnA/cDbgPfYfnXZP6ON9YuIiOe4Vldt/QNwLnCy7acbQdurJf1FW2oWERFdodVE8i7gadubASQ9D3ih7V/avrhttYuIiOe8VudIvkP1hcCGnUosIiImuFYTyQttP9U4KPs7tadKERHRTVpNJL+Q9NrGgaTZwNMjXB8RERNEq3MknwG+Ianx5t0pwAfaUqOIiOgqLSUS29+X9CrglVS/A/JD279ua80iIqIrbMtLG18H9JYyh0jC9kVtqVVERHSNlhKJpIuB/YC7qV7zDtWvESaRRERMcK32SOYAB+SHoyIiYrBWE8kK4GVUv60eEeNM7+LrOvLclacd1ZHnxuhqNZHsAdwn6XbKb6sD2H5vW2oVERFdo9VEcmo7KxEREd2r1d8juQlYCTy/7H8fuHOkMpLOl7RO0oqm2EslLZf0o/J3t6ZzJ0nql/SApCOb4rMl3VPOnanyY++SdpB0eYnfJql3WxoeERGjo9XXyH8MuBL4pxKaCnzzWYpdAMwdFFsM3GB7JnBDOUbSAcB84MBS5ixJk0qZs4FFwMyyNe55PPC47f2pXmV/eittiYiI0dXqK1JOBN4IPAm//ZGrPUcqYPtm4LFB4aOBC8v+hcC8pvhltjfafhjoBw6VNAXYxfYtZcXYRYPKNO51JXBEo7cSERFjp9VEstH2M40DSZOpvkeyrfayvQag/G0ko6nAqqbrBkpsatkfHN+ijO1NwBPA7kM9VNIiSX2S+tavX78d1Y6IiOG0mkhuknQysGP5rfZvAP88ivUYqifhEeIjldk6aJ9je47tOT09PdtZxYiIGEqriWQxsB64B/g4cD3V77dvq7VluIryd12JDwDTm66bBqwu8WlDxLcoU3pIL2HrobSIiGizVldt/cb2ubaPtX1M2d+eoa2lwMKyvxC4tik+v6zE2odqUv32Mvy1QdJhZf5jwaAyjXsdA9yYb95HRIy9Vt+19TBDDBvZ3neEMpcChwN7SBoATgFOA66QdDzwCHBsuc+9kq4A7gM2ASc2ftYXOIFqBdiOwLKyAZwHXCypn6onMr+VtkRExOjalndtNbyQKgG8dKQCto8b5tQRw1y/BFgyRLwPOGiI+K9KPSIiooNaHdr6WdP2qO2/B97W3qpFREQ3aHVo67VNh8+j6qG8uC01ioiIrtLq0Nbnm/Y3Ub0u5f2jXpuIiOg6rf7U7lvbXZGIiOhOrQ5tfXak87a/MDrViYiIbrMtq7ZeR/XdDYD3ADez5WtNIiJiAtqWH7Z6re0NAJJOBb5h+6PtqlhERHSHVl+RMgN4pun4GaB31GsTERFdp9UeycXA7ZKuofqG+/uoXukeERETXKurtpZIWga8uYQ+bPuu9lUrIiK6RatDWwA7AU/a/iIwUF6uGBERE1yrP7V7CvBnwEkl9Hzga+2qVEREdI9WeyTvA94L/ALA9mryipSIiKD1RPJM+a0PA0jauX1VioiIbtJqIrlC0j8Bu0r6GPAd4Nz2VSsiIrrFs67aKr9MeDnwKuBJ4JXAX9pe3ua6RUREF3jWRGLbkr5pezaQ5BEREVtodWjrVkmva2tNIiKiK7X6zfa3Ap+QtJJq5ZaoOisHt6tiERHRHUZMJJJm2H4EeOdoPVDSK6nmXBr2Bf4S2BX4GLC+xE+2fX0pcxJwPLAZ+JTtfy3x2cAFwI7A9cCny+qyiIgYI882tPVNANs/Br5g+8fN2/Y80PYDtmfZngXMBn4JXFNOn9E415REDgDmAwcCc4GzJE0q158NLAJmlm3u9tQpIiK237MlEjXt79uG5x8BPPgsSelo4DLbG20/DPQDh0qaAuxi+5bSC7kImNeGOkZExAieLZF4mP3RMh+4tOn4k5J+IOl8SbuV2FS2/AGtgRKbWvYHx7ciaZGkPkl969evH+qSiIjYTs+WSF4j6UlJG4CDy/6TkjZIerLOgyW9gOq1K98oobOB/YBZwBrg841LhyjuEeJbB+1zbM+xPaenp6dOtSMiYpARJ9ttTxrpfE3vBO60vbY8a23jhKRzgW+VwwFgelO5acDqEp82RDwiIsbQtrxGfrQdR9OwVpnzaHgfsKLsLwXmS9qhvLp+JnC77TXABkmHlW/fLwCuHZuqR0REQ6vfIxlVknYC3g58vCn8t5JmUQ1PrWycs32vpCuA+4BNwIm2N5cyJ/C75b/LyhYREWOoI4nE9i+B3QfFPjTC9UuAJUPE+4CDRr2CERHRsk4ObUVExDiQRBIREbUkkURERC1JJBERUUsSSURE1JJEEhERtSSRRERELUkkERFRSxJJRETUkkQSERG1JJFEREQtSSQREVFLEklERNSSRBIREbUkkURERC1JJBERUUsSSURE1NKRX0iMiADoXXxdx5698rSjOvbs8aYjPRJJKyXdI+luSX0l9lJJyyX9qPzdren6kyT1S3pA0pFN8dnlPv2SzpSkTrQnImIi6+TQ1lttz7I9pxwvBm6wPRO4oRwj6QBgPnAgMBc4S9KkUuZsYBEws2xzx7D+ERHBc2uO5GjgwrJ/ITCvKX6Z7Y22Hwb6gUMlTQF2sX2LbQMXNZWJiIgx0qlEYuDbku6QtKjE9rK9BqD83bPEpwKrmsoOlNjUsj84HhERY6hTk+1vtL1a0p7Ackk/HOHaoeY9PEJ86xtUyWoRwIwZM7a1rhERMYKO9Ehsry5/1wHXAIcCa8twFeXvunL5ADC9qfg0YHWJTxsiPtTzzrE9x/acnp6e0WxKRMSEN+aJRNLOkl7c2AfeAawAlgILy2ULgWvL/lJgvqQdJO1DNal+exn+2iDpsLJaa0FTmYiIGCOdGNraC7imrNSdDFxi+18kfR+4QtLxwCPAsQC275V0BXAfsAk40fbmcq8TgAuAHYFlZYuIiDE05onE9kPAa4aI/ww4YpgyS4AlQ8T7gINGu44REdG659Ly34iI6EJJJBERUUsSSURE1JJEEhERtSSRRERELUkkERFRSxJJRETUkkQSERG1JJFEREQtSSQREVFLEklERNSSRBIREbUkkURERC1JJBERUUsSSURE1JJEEhERtSSRRERELUkkERFRSxJJRETUMuaJRNJ0Sd+VdL+keyV9usRPlfSopLvL9q6mMidJ6pf0gKQjm+KzJd1Tzp0pSWPdnoiIiW5yB565Cfic7TslvRi4Q9Lycu4M23/XfLGkA4D5wIHA3sB3JL3C9mbgbGARcCtwPTAXWDZG7YiICDrQI7G9xvadZX8DcD8wdYQiRwOX2d5o+2GgHzhU0hRgF9u32DZwETCvvbWPiIjBOjpHIqkXOAS4rYQ+KekHks6XtFuJTQVWNRUbKLGpZX9wfKjnLJLUJ6lv/fr1o9mEiIgJr2OJRNKLgKuAz9h+kmqYaj9gFrAG+Hzj0iGKe4T41kH7HNtzbM/p6empW/WIiGjSkUQi6flUSeTrtq8GsL3W9mbbvwHOBQ4tlw8A05uKTwNWl/i0IeIRETGGOrFqS8B5wP22v9AUn9J02fuAFWV/KTBf0g6S9gFmArfbXgNskHRYuecC4NoxaURERPxWJ1ZtvRH4EHCPpLtL7GTgOEmzqIanVgIfB7B9r6QrgPuoVnydWFZsAZwAXADsSLVaKyu2IqIlvYuv68hzV552VEee205jnkhsf4+h5zeuH6HMEmDJEPE+4KDRq11ERGyrfLM9IiJqSSKJiIhakkgiIqKWJJKIiKgliSQiImpJIomIiFqSSCIiopYkkoiIqCWJJCIiakkiiYiIWpJIIiKiliSSiIioJYkkIiJqSSKJiIhakkgiIqKWJJKIiKgliSQiImpJIomIiFqSSCIiopYx/8320SZpLvBFYBLwFdundbhKERHD6l18XceevfK0o9py367ukUiaBPwj8E7gAOA4SQd0tlYRERNLVycS4FCg3/ZDtp8BLgOO7nCdIiImlG4f2poKrGo6HgBeP/giSYuAReXwKUkPbOfz9gB+up1lu1HaO35NpLZC2guATq91z5cPd6LbE4mGiHmrgH0OcE7th0l9tufUvU+3SHvHr4nUVkh7263bh7YGgOlNx9OA1R2qS0TEhNTtieT7wExJ+0h6ATAfWNrhOkVETChdPbRle5OkTwL/SrX893zb97bxkbWHx7pM2jt+TaS2QtrbVrK3mlKIiIhoWbcPbUVERIclkURERC1JJC2SNFfSA5L6JS3udH1Gk6Tpkr4r6X5J90r6dIm/VNJyST8qf3frdF1Hk6RJku6S9K1yPG7bK2lXSVdK+mH59/yG8dpeSf+t/O94haRLJb1wPLVV0vmS1kla0RQbtn2STiqfWw9IOrIddUoiacEEeBXLJuBztl8NHAacWNq3GLjB9kzghnI8nnwauL/peDy394vAv9h+FfAaqnaPu/ZKmgp8Cphj+yCqRTjzGV9tvQCYOyg2ZPvK/4/nAweWMmeVz7NRlUTSmnH9Khbba2zfWfY3UH3ITKVq44XlsguBeR2pYBtImgYcBXylKTwu2ytpF+AtwHkAtp+x/XPGaXupVqPuKGkysBPVd8vGTVtt3ww8Nig8XPuOBi6zvdH2w0A/1efZqEoiac1Qr2KZ2qG6tJWkXuAQ4DZgL9troEo2wJ4drNpo+3vgT4HfNMXGa3v3BdYDXy1DeV+RtDPjsL22HwX+DngEWAM8YfvbjMO2DjJc+8bksyuJpDUtvYql20l6EXAV8BnbT3a6Pu0i6d3AOtt3dLouY2Qy8FrgbNuHAL+gu4d2hlXmBo4G9gH2BnaW9EedrVVHjclnVxJJa8b9q1gkPZ8qiXzd9tUlvFbSlHJ+CrCuU/UbZW8E3itpJdUw5dskfY3x294BYMD2beX4SqrEMh7b+wfAw7bX2/41cDXwe4zPtjYbrn1j8tmVRNKacf0qFkmiGj+/3/YXmk4tBRaW/YXAtWNdt3awfZLtabZ7qf5d3mj7jxi/7f0JsErSK0voCOA+xmd7HwEOk7RT+d/1EVRzfuOxrc2Ga99SYL6kHSTtA8wEbh/th+eb7S2S9C6qcfXGq1iWdLZGo0fSm4B/A+7hd3MGJ1PNk1wBzKD6P+ixtgdP8nU1SYcD/932uyXtzjhtr6RZVAsLXgA8BHyY6j8kx117Jf0V8AGq1Yh3AR8FXsQ4aaukS4HDqV4VvxY4Bfgmw7RP0p8DH6H65/EZ28tGvU5JJBERUUeGtiIiopYkkoiIqCWJJCIiakkiiYiIWpJIIiKiliSSiIioJYkkIiJq+f/nT8z8wtiQYAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.age.plot(kind='hist',bins=10)\n",
    "plt.title('Age Histogram');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "pd.crosstab(df.age, df.gender).plot(style=['or','ob'], alpha=0.8);\n",
    "plt.title('Age Frequency by Gender')\n",
    "plt.ylabel('Frequency');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Looking at the Age histogram with a higher number of bins (more granularity), we can see there are basically three age groups with higher frequency than the others: there is a peak for very young patients; another one at around 30-35 years; and the last one around 45-55 years.\n",
    "\n",
    "The second peak seems to be formed due to a high frequency of females, whereas the third peak has contribution of both male and female, given their proportions in the dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# interval of delta days to be plotted\n",
    "df['delta_weeks'] = pd.cut(df.delta_date, [pd.Timedelta(i, 'D') for i in range(0,90,7)])\n",
    "delta_proportion_weeks = df.groupby('delta_weeks').size() / df.shape[0]\n",
    "\n",
    "delta_proportion_weeks.cumsum().plot(style='k--', label='CumSum')\n",
    "\n",
    "delta_proportion_weeks.plot(kind='bar', color='xkcd:blue', alpha=0.8, label='_No')\n",
    "plt.title('Proportion of Scheduling [in Weeks]')\n",
    "plt.xticks(range(12), ['Week {}'.format(x) for x in range(1,13)])\n",
    "plt.ylabel('Proportion of Appointments Scheduled')\n",
    "plt.legend();\n",
    "\n",
    "# print(delta_proportion_weeks.round(2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The majority of appointments are scheduled 5 weeks in advance (around 60%), being almost half of these made within 1 week from the actual appointment. Let's further investigate what it looks like for the first 2 weeks:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "delta_days\n",
      "(0 days 00:00:00, 1 days 00:00:00]      0.05\n",
      "(1 days 00:00:00, 2 days 00:00:00]      0.06\n",
      "(2 days 00:00:00, 3 days 00:00:00]      0.02\n",
      "(3 days 00:00:00, 4 days 00:00:00]      0.05\n",
      "(4 days 00:00:00, 5 days 00:00:00]      0.03\n",
      "(5 days 00:00:00, 6 days 00:00:00]      0.04\n",
      "(6 days 00:00:00, 7 days 00:00:00]      0.04\n",
      "(7 days 00:00:00, 8 days 00:00:00]      0.02\n",
      "(8 days 00:00:00, 9 days 00:00:00]      0.01\n",
      "(9 days 00:00:00, 10 days 00:00:00]     0.01\n",
      "(10 days 00:00:00, 11 days 00:00:00]    0.01\n",
      "(11 days 00:00:00, 12 days 00:00:00]    0.01\n",
      "(12 days 00:00:00, 13 days 00:00:00]    0.02\n",
      "(13 days 00:00:00, 14 days 00:00:00]    0.03\n",
      "dtype: float64\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# interval of delta days to be plotted: 14 days\n",
    "df['delta_days'] = pd.cut(df.delta_date, [pd.Timedelta(i, 'D') for i in range(0,15)])\n",
    "\n",
    "delta_proportion_days = df.groupby('delta_days').size() / df.shape[0]\n",
    "\n",
    "delta_proportion_days.plot(kind='bar', color='xkcd:blue', alpha=0.8)\n",
    "plt.title('Proportion of Scheduling [first 14 days]')\n",
    "plt.xticks(range(14), ['Day {}'.format(x) for x in range(1,15)])\n",
    "plt.ylabel('Proportion of Appointments Scheduled')\n",
    "\n",
    "print(delta_proportion_days.round(2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There is no clear pattern for the proportion of appointments scheduled within 14 days in advance.\n",
    "Same day appointments - assumed to be \"Emergency\" ones - represent only 5% of the sample."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Research Questions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since we are interested in investigating the correlation between a specific variable and the final outcome (show vs no-show), it is worth writing a function that will be used multiple times.\n",
    "\n",
    "The function will cross reference the desired variable against the final outcome (column Attended), normalizing the result between Yes-Attended and No-Attended and plotting a bar chart with the respective title and labels:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "def proportion_attendance(feature):\n",
    "    pd.crosstab(df[feature], df.attended, normalize='columns').plot(kind='bar', alpha=0.85)\n",
    "    plt.xlabel('Feature: {}'.format(feature.replace('_',' ').title()))\n",
    "    plt.ylabel('Proportion')\n",
    "    plt.title('Proportion of Attendance Rate by {}'.format(feature.replace('_',' ').title()))\n",
    "    plt.margins(y=0.1)\n",
    "    return plt # this allows for eventual further formatting tweaks"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Q1) Does gender impact the attendance?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "proportion_attendance('gender');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There is no clear evidence that gender have any influence on whether or not the patient is going to show up for their scheduled appointment."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Q2) Is there a specific age group with higher attendance rate?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since the age distribution is so spread out, let's transform this numeric variable into age groups - representing $10^{th} $ percentile each - to further investigate this question:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df['age_groups'] = pd.qcut(df.age, 10)\n",
    "proportion_attendance('age_groups');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It looks like age does have some influence over attendance rate. We can clearly see that patients between 5 and 44 years old seem to be less likely to show up, whereas patients older than 45 seem to be more likely to attend their scheduled appointment. It is also possible to see that patients younger than 5 years old tend to be slightly more likely to attend.\n",
    "\n",
    "Since the age groups present a distinctively proportion of attendance rate, we can create dummy variables (Hot-One encoded) to capture their influence in the classification model. We will tackle this step later."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Q3) Does date difference between the scheduled and the appointment day impact the attendance rate? \n",
    "\n",
    "Here we are particularly interested in investigating if there is evidence of any pattern that influence attendance rate. For example, one might wonder whether same day appointments have a higher attendance rate, since they could be viewed as surrogate for emergency visits?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "proportion_attendance('delta_days')\n",
    "plt.xticks(range(14), ['Day {}'.format(i) for i in range(1,15)]);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It looks like the difference between appointment and scheduled dates have some influence over the attendance rate. We can clearly see that appointments scheduled within 4 days in advance seem to be more likely to show up: these could be interpreted as \"same week appointments\".\n",
    "\n",
    "In contrast, those scheduled longer than 7 days in advance tend to be less likely to attend.\n",
    "\n",
    "Since the attendance rate doesn't vary significantly for all delta days, perhaps it would be more meaningful to create 2 dummy variables, representing: appointments scheduled within 4 days in advance, and longer than 7 days in advance."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "min      0 days 00:00:00\n",
      "max      4 days 17:24:06\n",
      "count              61801\n",
      "Name: delta_date, dtype: object\n",
      "min        7 days 03:36:35\n",
      "max      178 days 13:19:01\n",
      "count                39772\n",
      "Name: delta_date, dtype: object\n"
     ]
    }
   ],
   "source": [
    "# Checking for correct filtering\n",
    "print(df[df.delta_date < pd.Timedelta(5,'D')].delta_date.aggregate([min,max, 'count']))\n",
    "print(df[df.delta_date > pd.Timedelta(7,'D')].delta_date.aggregate([min,max, 'count']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create 2 new features to represent if appointment scheduled <= 4 and > 7 days in advance\n",
    "df['reg_up_4days'] = np.where( df.delta_date < pd.Timedelta(5,'D') , 1, 0)\n",
    "df['reg_more_7days'] = np.where( df.delta_date > pd.Timedelta(8,'D') , 1, 0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Q4) Are there specific neighborhoods associated with higher No-Show?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>neighborhood</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>PARQUE INDUSTRIAL</th>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ILHA DO BOI</th>\n",
       "      <td>35</td>\n",
       "      <td>0.914286</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AEROPORTO</th>\n",
       "      <td>8</td>\n",
       "      <td>0.875000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MÁRIO CYPRESTE</th>\n",
       "      <td>371</td>\n",
       "      <td>0.854447</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SOLON BORGES</th>\n",
       "      <td>469</td>\n",
       "      <td>0.852878</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   count      mean\n",
       "neighborhood                      \n",
       "PARQUE INDUSTRIAL      1  1.000000\n",
       "ILHA DO BOI           35  0.914286\n",
       "AEROPORTO              8  0.875000\n",
       "MÁRIO CYPRESTE       371  0.854447\n",
       "SOLON BORGES         469  0.852878"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.pivot_table('attended', index='neighborhood', aggfunc=('mean', 'count'))\\\n",
    "    .sort_values(by='mean', ascending=False).head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "At first glance, we can identify some neighborhoods with high attendance rate... but it is important to take into consideration the representativeness of that data. It would be biased to infer the attendance rate of specific neighborhoods based only on the observations of few patients from that specific location (such as Parque Industrial with 100% attendance rate, but only 1 patient).\n",
    "\n",
    "Therefore, for this particular research questions, let's work only with neighborhoods where the count is equal or greater than the median count of all neighborhoods. Let's call this subset $Neighborhood^*$:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "# count of all neighborhoods\n",
    "neighb_count = df.groupby('neighborhood').size()\n",
    "\n",
    "# subset, keeping only counts >= the median\n",
    "neighb_subset = neighb_count[neighb_count >= neighb_count.median()]\n",
    "\n",
    "# new dataframe, filtered to the previous subset\n",
    "df2 = df.query('neighborhood in {}'.format(list(neighb_subset.index)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA4UAAAGyCAYAAABTMGGcAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAACn7ElEQVR4nOzdebx1c93/8dfblTnjbUhmopKQTNGvwXCHEk24JDSpUFQq0TwX3UmJJkkpaRAVTZRyu4XMY8lcZKg0qITP74/Pdzvr7GufffZae+3rnHOd9/Px2I9z9l57ffZ3T2uv7/T5KiIwMzMzMzOz2WmhqS6AmZmZmZmZTR1XCs3MzMzMzGYxVwrNzMzMzMxmMVcKzczMzMzMZjFXCs3MzMzMzGYxVwrNzMzMzMxmMVcKzWzKSTpR0gdajrmfpPPajFnjsdeSFJIeVa6fJWnfyvYPSLpH0p3l+gsk3Sbp75KeMhVlHlZ5vo+b6nLMFlP5+bb6JB0u6QsD3vc9kr7aZ/vNkrZvqVzPknR7G7EGeKzWym1m7XOl0MzmG0k/l/RnSYtOdVmqRn2CHRE7RcSXy2OtDrwZ2CAiHlPuchRwUEQ8OiIuHVU5pquZUqGUtGSpuJ/Zdfu4RoBy26yqtFVeg7+Xy82SDquxf+sNQ20qx65/le9v57btJd08yP4R8aGIeNXICmhmNiRXCs1svpC0FvD/gACeP7WlmVJrAvdGxF1dt13dJFi1ImIj92Lg38B/S1plqgszTS0bEY8mX6t3StphqgtUl6Q5E2z6B/DO+VmWUfKxw8yqXCk0s/llH+AC4ERg3x7bV5D0E0l/k3SupDUBlD4h6S5J90m6QtKGZdsykk6SdLekWyS9Q9I8x7UJenJ+LulVkp4IHA88rfRw/KVsX1TSUZJulfRHScdLWrzXE5M0p9z3Hkk3As/t2t55rO2BnwCPLY/1dUl/B+YAl0v6Xbn/YyV9uzyvmyS9oRLrPZK+Jemrkv4K7Fdehy9KukPS78vw1Dnl/vtJOq+U788l3k6VeMtL+pKkP5Tt361se56kyyT9RdL5kjaa4L3t2FnSjeV1OLL6Xkh6haRry2P8qPL+/qLc5fLymuxR3v8Xle1PL+/dzuX69pIumyxu2faE8pn6k6TrJe1e2XaipGMl/aB85n4lad1Jnt++5GflCuCllds7z+Ev5Tk8jZqfKZVhfJLeXD7rd0h6eaW8/yXpDEl/lXQhMK6skj6pHIL8V0m/lvT/KtveI+nU8l35m6SrJW1W2b66pO+Uz9u9kj49yOvbT0RcTDZ0bFKJ9U1Jdyq/x7+Q9KRy+/7l9Xxreb2+V26f8HvQrbyfx6vHMaRsn+yzcJykMyX9A3j2BA9zDDBXE/Rq9yuvuoaEStpHecy6V9I7Ne/QykUmer+KzSVdU96XL0larBL71ZJuKM/1DEmPrWwLSQdK+i3w28rtE33uJjzGSlqoXL+l7HuSpGUq+76s8hyPmOA1NbPpIiJ88cUXX0Z+AW4ADgCeCvwHWLmy7UTgb8AzgEWBTwLnlW3PAX4NLAsIeCKwStl2EnA6sBSwFvAb4JVl236VGGuRPZSPqjzmz4FXdd+3sv1o4Axg+RL/e8CHJ3hurwWuA1Yv9/9Z9fG6HutZwO1d+wfwuPL/QuX5vgtYBFgHuBF4Ttn+nvL67VbuuzjwXeCzwJLASsCFwGsqz+0/wKvJyufrgD8AKtt/AHwDWA5YGHhmuX1T4C5gy7LfvsDNwKITvAZRnvfywBrlveg8593I9/+JwKOAdwDn93r+5fr7gE+V/w8Hfgd8tLLtk5PFLa/FbcDLy7ZNgXuAJ1U+c38CtijbTwZO6fP5XQN4GNiAHP57RWXbWsz7+dqPGp8p8nPxYHl+CwM7A/cDy5XtpwCnlue1IfD7anxgb+C/ynN5M3AnsFjlM/OvEnMO8GHggrJtDnA58IkSezHg6YO8b13PbdxrAGxVyv+Cyn1eUZ73ouW1uKzrGPCByvW+34Mej38iEx9DBvks3AdsUx53sR7xfw68Cvgf4Kvltu2Bm2t8bzv7bQD8HXh6ue9R5Hd0+8ner7L9ZuAqxo43/9t57YBty3PbtLwOnwJ+0fVd+0nZb3Em/9z1O8a+gvx8rAM8GvgO8JWu59h5P/6nPM72U/H744svvkx+mfIC+OKLLwv+pZz8/AdYoVy/DnhjZfuJVE7IywnGQ+WkZ9tyIrIVsFDlPnPIoXwbVG57DfDz8v9+NKwUkpXPfwDrVm57GnDTBM/vHOC1lev/TfNK4ZbArV3b3w58qfz/nq6TvJXL67B45ba5wM8qz+2GyrYlyuM9BliFrOgs1+M5HQe8v+u26ymVxh73D2DHyvUDgLPL/2d1TiTL9YXIE881u59/ub4dpdIF/JA8Ge9UYs4FXjhZXGAP4JddZfws8O7KZ+4LlW07A9f1+Qy/g1KJAR5Lfj6f0ufzVeszVT4X/+yKcRf5uZ9Dfn+eUNn2IboqnV3l/TOwceUz89PKtg2Af1bKcHf1cSv36/u+dd238xr8pTyPICs7mqB8y5b7LFN5P6qVwr7fgx7xTmTiY8ggn4WTJnotq99hYEWyAvkkxlcKB/nediqF7wK+3vWdfIDxlcKe71e5fjPjjzc7A78r/38R+FjX6/AfYK3Kd23byvbJPnf9jrFnAwdUtj2+PNajynOsvh9LVp+jL774Mv0uHj5qZvPDvsCPI+Kecv1rzDuE9LbOPxHxd7IX57ERcQ7waeBY4I+SPidpaWAFspX9lkqMW4BVWyjviuSJ2q+VQyf/QlZOVpzg/o+tlr+rTHWtSQ4v/UvlsQ8nK38dt3Xdf2Hgjsr9P0v2GHbc2fknIu4v/z6aPGH+U0T8eYJyvLmrHKuTz3Ui3a9B575rAp+sxPkTWUma6L36P2B9SSuTww9PAlaXtALZs9cZrtkv7prAll3lfylZGe64s/L//eRrMpF9yN5EIuIPZOW0+zPczyCfqXsj4sEeZVqRPNGe8DNWhv9dW4Zm/gVYhvyOdHQ/18WUw6lXB27petyOuu8b5TEfDRxKVjgWLuWbI+kjkn6nHPZ8c+X+vQzyPejW8xjCYJ+F6ms7oYi4mzwevW+I8o47XpTv5L1d95no/epV3up37bFUPhvldbiX8e9Z93Od6HM32TH2sT22PYp8zt3P8R89nqOZTSOeZGxmI6WcM7U7MEdlCQZyONGykjaOiMvLbdWsfo8mhzf9ASAijgGOkbQSOYTuLYwNo1wTuKbsugY5rK7bP8rfJYC/lv+rJ4TRdf97yNbzJ0VEr3jd7qiWv5SjqdvI3qP1+tynWt7byNb8FSY4sZ/ssZaXtGxE/KXHtg9GxAdrxFudsYQ5a1Dev0qskwcJEhH3S/o1cDBwVUQ8IOl84E1kj0incWHCuGU+2bkRMXSiE0lbA+sBb5f05nLzUsCTJB3KvJ8fetxW9zNVdTc59G51spcdKp8x5fzBt5E9rFdHxMOS/kxW4CZzG7CGpEf1+PzUet86IuIh4OOSXkD2GB8N7AXsSuldIyut1TJ2v16DfA+6TXQMuY3JPwu93sOJHEkODb2wclud8t5B9qp1yro4OfS3ju7jTee79gfymNiJvWSJXf3MDfpc76H/MXbcY5VtDwJ/JJ/jEyvlWIL6z9HM5iP3FJrZqO1GDuPagOz12YQ8Wfgl2fvSsbMyqcgiwPuBX0XEbZI2l7SlpIXJyt2/gIfKieepwAclLVUqAW8C5lnfq7Tu/x7Yu/RYvILxiTr+CKxWHpuIeBj4PPCJUhFF0qqSnjPBczwVeIOk1SQtBwycir+HC4G/SnqbpMVLeTeUtHmvO0fEHcCPyZPwpUvyh3UlPXOyByr7ngV8RtJykhaW9Iyy+fPAa8trL+VyDM+VtFSfkG8pcVYnK3TfKLcfT1aoOolFlpH0ksp+fyTnJVWdCxxU/kIO36tenyzu98nexpeV57Vw+Sw9kfr2JedhVT/DG5KNDDuRlbaHu57DsJ+pR5TP+neA90haQtIGjO+lXIo8Gb8beJSkdwFLD/jcLiRP4D9S3uPFJG1Ttk32vk3mI2TymMVKGf9N9hYtQQ5/rer+DNT6HhQ9jyG0+1mgNKB8HHhrw/J+C9hF0talrO9lsAp81YHleLM82SPZ+a59DXi5pE2US/98iHwdbq4ZnwGOsV8H3ihp7VIJ/xDwjdK48C3geZX34334nNNsWvMX1MxGbV9yXs2tEXFn50IOwXppZUjU14B3k0O+nspYdselyZPpP5PDk+4l5yoBvJ6sKN4InFdinDBBOV5N9jDeS84HOr+y7Ryyh+tOSZ1eqLeRSRQuUA53+ymV1v0unwd+RCbsuIQ8gW+knIjtQlY8biJb679A9qxMZB9ymNc15Ov0LXK+4CBeRvYGXEfOJTqklONi8jX7dIl5AzlPrp/TyWQbl5EJbL5YYp0GfBQ4pbyWV5GVqY73AF8uw+46WSHPJSsSv5jget+4EfE3cm7nnmSPxp3lvrXWyCwVmt3JxDd3Vi43AV8B9i3D/z4I/G95Dlsx/Geq20HkkL47yTlwX6ps+xFZuf8N+R35F4MPh+x83h4H3ArcTs7BG+R9m8wPyM/Oq8khwLeQjTPXkJmIq74IbFBev+82/B70PIa09Vno8kmysYvyGAOXNyKuJo9dp5AV8r+R371/13j8r5GNQTeWywdK7LPJZTO+XWKvSz7vpvodY08gvwO/IJ/zv8r9O8/xwHL/O8jPwe1DlMPMRqyTfc7MzMxsRpJ0IpnA6R1TXZa6Si/bX4D1SmODmdl8555CMzMzs/lI0i5lKPCS5MiHKxlLvmNmNt+5UmhmZmY2f+1KDmX9A5nEaM/w0C0zm0IePmpmZmZmZjaLuafQzMzMzMxsFnOl0MzMzMzMbBabFYvXr7DCCrHWWmtNdTFslG65Jf+uuWb/+5mZmZmZzUK//vWv74mIFXttmxWVwrXWWouLL754qotho/T2t+ffD394asthZmZmZjYNSbplom0jHT4qaUdJ10u6QdJhPbZL0jFl+xWSNq1sO0HSXZKumiD2oZJC0gqjfA42Q3z4w64QmpmZmZk1MLJKoaQ5wLHATsAGwFxJG3TdbScyFfN6wP7AcZVtJwI7ThB7dWAH4NZ2S21mZmZmZja7jLKncAvghoi4MSIeAE4h1+Wp2hU4KdIFwLKSVgGIiF8Af5og9ieAtwJeT8PSy1+eFzMzMzMzq2WUcwpXBW6rXL8d2HKA+6wK3DFRUEnPB34fEZdLaqmoNuOtvvpUl8DMzMzMbEYaZaWwV42tu2dvkPuM3VlaAjgC+O9JH1zanxySyhprrDHZ3W2me9/7proEZmZmZmYz0iiHj94OVLtvVgP+0OA+VesCawOXS7q53P8SSY/pvmNEfC4iNouIzVZcsWfmVTMzMzMzs1lvlJXCi4D1JK0taRFgT+CMrvucAexTspBuBdwXERMOHY2IKyNipYhYKyLWIiuVm0bEnSN6DjZT7L13XszMzMzMrJaRDR+NiAclHQT8CJgDnBARV0t6bdl+PHAmsDNwA3A/8EimEElfB54FrCDpduDdEfHFUZXXZrjHP36qS2BmZmZmNiMpYsFP4LnZZpuFF683MzMzM7PZStKvI2KzXttGuni9mZmZmZmZTW+uFNqCYc8982JmZmZmZrWMckmKaW3bA24d6H7nfMbLWcwIm2wy1SUwMzMzM5uRZm2l0BYwhx021SUwMzMzM5uRPHzUzMzMzMxsFnOl0BYML3pRXszMzMzMrBYPH7UFw9OeNtUlMDMzMzObkVwptAXDoYdOdQnMzMzMzGYkDx81MzMzMzObxVwptAXD85+fFzMzMzMzq8XDR23BsN12U10CMzMzM7MZyZVCWzAcfPBUl8DMzMzMbEby8FEzMzMzM7NZzJVCWzDstFNezMzMzMysFg8ftQXDLrtMdQnMzMzMzGYkVwptwXDAAVNdAjMzMzOzGcnDR83MzMzMzGYxVwptwbD99nkxMzMzM7NaPHzUFgx77DHVJTAzMzMzm5FcKbQFw6tfPdUlMDMzMzObkTx81MzMzMzMbBZzpdAWDM96Vl7MzMzMzKwWDx+1BcN++011CczMzMzMZiRXCm3B4EqhmZmZmVkjHj5qC4b//CcvZmZmZmZWS9+eQkl/nWR/AXdExPrtFcmsgR12yL8///mUFsPMzMzMbKaZbPjo7yLiKf3uIOnSFstj1syrXjXVJTAzMzMzm5EmqxS+aIAYg9zHbLT23nuqS2BmZmZmNiP1rRRGxI2d/yWtDGxerl4YEXd138dsytx/f/5dYompLYeZmZmZ2QwzUKIZSbsDFwIvAXYHfiXpxQPst6Ok6yXdIOmwHtsl6Ziy/QpJm1a2nSDpLklXde1zpKTryv1Pk7TsIM/BFnA775wXMzMzMzOrZdDso0cAm0fEvhGxD7AF8M5+O0iaAxwL7ARsAMyVtEHX3XYC1iuX/YHjKttOBHbsEfonwIYRsRHwG+DtAz4HW5C97nV5MTMzMzOzWgZdp3ChznDR4l4mr1BuAdzQGV4q6RRgV+Cayn12BU6KiAAukLSspFUi4o6I+IWktbqDRsSPK1cvACbtsbRZYI89proEZmZmZmYz0qCVwh9K+hHw9XJ9D+DMSfZZFbitcv12YMsB7rMqcMeA5XoF8I0B72sLsvvuy7/LLDO15TAzMzMzm2EGqhRGxFskvQjYhlyb8HMRcdoku6lXqAb36R1cOgJ4EDh5gu37k0NSWWONNQYJaTPZrrvmX69TaGZmZmZWy6A9hUTEt4Fv14h9O7B65fpqwB8a3GcekvYFngdsV4ae9irv54DPAWy22WYDVTRtBnvDG6a6BGZmZmZmM1LfSqGkm5i45y4iYt0+u18ErCdpbeD3wJ7AXl33OQM4qMw33BK4LyL6Dh2VtCPwNuCZEXF/v/vaLPLCF051CczMzMzMZqTJego367q+ELkkxaHApf12jIgHJR0E/AiYA5wQEVdLem3Zfjw5L3Fn4AbgfuDlnf0lfR14FrCCpNuBd0fEF4FPA4sCP5EEcEFEvHbyp2oLtHvuyb8rrDC15TAzMzMzm2EmW7z+XgBJCwEvA94CXAY8NyKu6bNrZ/8z6UpIUyqDnf8DOHCCfedOcPvjJntcm4VeXJLQek6hmZmZmVktkw0fXZjM8PlG4Dxg14j43fwomFktb37zVJfAzMzMzGxGmmz46E1khs+jgVuBjSVt3NkYEd8ZXdHMathll6kugZmZmZnZjDRZpfCnZKKZjculKgBXCm16uPPO/PuYx0xtOczMzMzMZpjJ5hTuN5/KYTacPffMv55TaGZmZmZWy8DrFHZI+n5EPG8UhTFr7LDDproEZmZmZmYzUu1KIbBq66UwG9aOO051CczMzMzMZqSFGuzTd31Csylx2215MTMzMzOzWpr0FL5Z0kYRcUXrpTFr6mUvy7+eU2hmZmZmVstAlUJJPweeX+5/GXC3pHMj4k2jK5pZDe94x1SXwMzMzMxsRhq0p3CZiPirpFcBX4qId0tyT6FNH9tvP9UlMDMzMzObkQadU/goSasAuwPfH2F5zJq58ca8mJmZmZlZLYP2FL4P+BHwvxFxkaR1gN+OrlhmNb3iFfnXcwrNzMzMzGoZqFIYEd8Evlm5fiPwolEVyqy29753qktgZmZmZjYjDTR8VNL6ks6WdFW5vpEkZ/aw6eOZz8yLmZmZmZnVMuicws8Dbwf+A1CWo9hzVIUyq+366/NiZmZmZma1DDqncImIuFBS9bYHR1Aes2Ze85r86zmFZmZmZma1DFopvEfSukAASHoxcMfISmVW14c+NNUlMDMzMzObkfpWCiW9Ffg4cCDwOeAJkn4P3ATsPfrimQ1o662nugRmZmZmZjPSZD2FawK/Bg6MiO0lLQksFBF/G33RzGq46qr8u+GGU1sOMzMzM7MZpm+lMCIOlLQp8ClJ1wHHAQ935hZGxCWjL6LZAA46KP96TqGZmZmZWS2TzimMiEskHQF8G3hkXmH5u+0Iy2Y2uCOPnOoSmJmZmZnNSJPNKVyJnFO4DrBtRFw+X0plVtfmm091CczMzMzMZqTJ1im8APgl8HRXCG1au+yyvJiZmZmZWS2TDR/dMiLuni8lMRvGIYfkX88pNDMzMzOrZbJEM64Q2sxw9NFTXQIzMzMzsxlp0MXrzaa3TTaZ6hKYmZmZmc1Ik80pHKesU2g2/Vx0UV7MzMzMzKyWgSqFkraWdA1wbbm+saTPjLRkZnW85S15MTMzMzOzWgYdPvoJ4DnAGQARcbmkZ4ysVGZ1ffrTU10CMzMzM7MZaeDhoxFxW9dND022j6QdJV0v6QZJh/XYLknHlO1XSNq0su0ESXdJuqprn+Ul/UTSb8vf5QZ9DrYA23DDvJiZmZmZWS2D9hTeJmlrICQtAryBMpR0IpLmAMcCOwC3AxdJOiMirqncbSdgvXLZEjiu/AU4Efg0cFJX6MOAsyPiI6WieRjwtgGfx8hse8CtA9/3nM+sMcKSzFLnn59/t956asthZmZmZjbDDNpT+FrgQGBVsoK3CXDAJPtsAdwQETdGxAPAKcCuXffZFTgp0gXAspJWAYiIXwB/6hF3V+DL5f8vA7sN+BxsQXb44XkxMzMzM7NaBu0pfHxEvLR6g6RtgP/ts8+qQHXI6e2M9QL2u8+qwB194q4cEXcARMQdklaapOw2G3z2s1NdAjMzMzOzGWnQnsJPDXhblXrcFg3u04ik/SVdLOniu+++u42QNp09/vF5MTMzMzOzWvr2FEp6GrA1sKKkN1U2LQ3MmST27cDqleurAX9ocJ9uf5S0SuklXAW4q9edIuJzwOcANttss1YqmjaNnXtu/n3mM6e2HGZmZmZmM8xkPYWLAI8mK49LVS5/BV48yb4XAetJWrskp9mTsqRFxRnAPiUL6VbAfZ2hoX2cAexb/t8XOH2S+9ts8O5358XMzMzMzGrp21MYEecC50o6MSJuqRM4Ih6UdBDwI7JX8YSIuFrSa8v244EzgZ2BG4D7gZd39pf0deBZwAqSbgfeHRFfBD4CnCrplcCtwEvqlMsWUCecMNUlMDMzMzObkQZNNHO/pCOBJwGLdW6MiG377RQRZ5IVv+ptx1f+DzKraa99505w+73AdgOW22aLddaZ6hKYmZmZmc1IgyaaORm4DlgbeC9wMzk81Gx6+OlP82JmZmZmZrUM2lP4XxHxRUkHV4aUnjvKgpnV8oEP5N/tt5/acpiZmZmZzTCDVgr/U/7eIem5ZIbQ1UZTJLMGvvKVqS6BmZmZmdmMNGil8AOSlgHeTK5PuDRwyKgKZVbb6qtPfh8zMzMzM5vHQJXCiPh++fc+4NkAkrYZVaHMavvhD/PvjjtObTnMzMzMzGaYyRavnwPsDqwK/DAirpL0POBwYHHgKaMvotkAPvKR/OtKoZmZmZlZLZP1FH4RWB24EDhG0i3A04DDIuK7Iy6b2eBOOWWqS2BmZmZmNiNNVincDNgoIh6WtBhwD/C4iLhz9EUzq+Exj5nqEpiZmZmZzUiTrVP4QEQ8DBAR/wJ+4wqhTUvf+15ezMzMzMyslsl6Cp8g6Yryv4B1y3UBEREbjbR0ZoP6+Mfz7y67TG05zMzMzMxmmMkqhU+cL6UwG9a3vjXVJTAzMzMzm5H6Vgoj4pb5VRCzoaywwlSXwMzMzMxsRhp08XqbAtsecOtA9zvnM2uMuCQzwHe+k39f+MKpLYeZmZmZ2QzjSqEtGI45Jv+6UmhmZmZmVsvAlUJJiwNrRMT1IyyPWTOnnz7VJTAzMzMzm5EmW5ICAEm7AJcBPyzXN5F0xgjLZVbPMsvkxczMzMzMahmoUgi8B9gC+AtARFwGrDWKApk18o1v5MXMzMzMzGoZdPjogxFxn6SRFsasseOOy7977DG15TAzMzMzm2EGrRReJWkvYI6k9YA3AOePrlhmNZ155lSXwMzMzMxsRhp0+OjrgScB/wa+BtwHHDKiMpnVt8QSeTEzMzMzs1oG6imMiPuBI8rFbPr56lfz7957T205zMzMzMxmmEGzj/5E0rKV68tJ+tHISmVW1xe+kBczMzMzM6tl0DmFK0TEXzpXIuLPklYaTZHMGvjJT6a6BGZmZmZmM9KgcwoflrRG54qkNYEYTZHMGlh44byYmZmZmVktg/YUHgGcJ+nccv0ZwP6jKZJZAyeemH/3228qS2FmZmZmNuMMmmjmh5I2BbYCBLwxIu4ZacnM6nCl0MzMzMyskUF7CgEWBf5U9tlAEhHxi9EUy6ymn/98qktgZmZmZjYjDVQplPRRYA/gauDhcnMArhSamZmZmZnNYIP2FO4GPD4i/j3Cspg19/nP599Xv3pqy2FmZmZmNsMMmn30RqB2akdJO0q6XtINkg7rsV2SjinbryjzFvvuK2kTSRdIukzSxZK2qFsuWwB94xt5MTMzMzOzWgbtKbwfuEzS2cAjvYUR8YaJdpA0BzgW2AG4HbhI0hkRcU3lbjsB65XLlsBxwJaT7Psx4L0RcZakncv1Zw34PGxB9dOfTnUJzMzMzMxmpEErhWeUSx1bADdExI0Akk4BdgWqlcJdgZMiIoALJC0raRVgrT77BrB02X8Z4A81y2VmZmZmZmbFoEtSfLlB7FWB2yrXbyd7Aye7z6qT7HsI8CNJR5HDX7duUDZb0HzmM/n3gAOmthxmZmZmZjPMQHMKJa0n6VuSrpF0Y+cy2W49bosB79Nv39eR6ySuDrwR+OIEZd6/zDm8+O67756kqDbjfe97eTEzMzMzs1oGTTTzJXK+34PAs4GTgK9Mss/twOqV66sx71DPie7Tb999ge+U/79JDlOdR0R8LiI2i4jNVlxxxUmKajPeWWflxczMzMzMahm0Urh4RJwNKCJuiYj3ANtOss9FwHqS1pa0CLAn885LPAPYp2Qh3Qq4LyLumGTfPwDPLP9vC/x2wOdgZmZmZmZmXQZNNPMvSQsBv5V0EPB7YKV+O0TEg+W+PwLmACdExNWSXlu2Hw+cCewM3EBmOH15v31L6FcDn5T0KOBfwP4DP1tbcH3yk/n34IOnthxmZmZmZjPMoJXCQ4AlgDcA7yd76PadbKeIOJOs+FVvO77yfwAHDrpvuf084KkDlttmi7PPzr+uFJqZmZmZ1TJo9tGLyr9/p/TmmU0rZ9RdMcXMzMzMzGCSSqGk7zFvxtBHRMTzWy+RmZmZmZmZzTeT9RQeVf6+EHgM8NVyfS5w84jKZCOy7QG3Dnzfcz6zxghLMgJHlY/qoYdObTnMzMzMzGaYvpXCiDgXQNL7I+IZlU3fk/SLkZbMrI7/+7+pLoGZmZmZ2Yw0aKKZFSWtExE3AkhaG/DifzZ9fPvbU10CMzMzM7MZadBK4RuBn0u6sVxfC3jNSEpkZmZmZmZm882g2Ud/KGk94Anlpusi4t+jK5ZZTR/5SP497LCpLYeZmZmZ2QwzaE8h5NqAa5V9NpZERJw0klKZ1XXZZVNdAjMzMzOzGWmgSqGkrwDrApcBD5WbA3Cl0KaHU06Z6hKYmZmZmc1Ig/YUbgZsEBETrlloZmZmZmZmM89CA97vKnKdQrPp6f3vz4uZmZmZmdUyaE/hCsA1ki4EHkkwExHPH0mpbMbY9oBbB7rfOZ9ZY7QFuf760cY3MzMzM1tADVopfM8oC2E2tK9+dapLYGZmZmY2Iw26JMW5ktYE1ouIn0paApgz2qKZmZmZmZnZqA00p1DSq4FvAZ8tN60KfHdEZTKr713vyouZmZmZmdUy6PDRA4EtgF8BRMRvJa00slKZ1XXbbVNdAjMzMzOzGWnQSuG/I+IBSQBIehS5TqHZ9PClL011CczMzMzMZqRBl6Q4V9LhwOKSdgC+CXxvdMUyMzMzMzOz+WHQSuFhwN3AlcBrgDMj4oiRlcqsrre/PS9mZmZmZlbLoMNHXx8RnwQ+37lB0sHlNrOpd++9U10CMzMzM7MZadCewn173LZfi+UwG87nPpcXMzMzMzOrpW9PoaS5wF7A2pLOqGxaCnDXjLVu2wNuHeh+53xmjRGXxMzMzMxsdphs+Oj5wB3ACsDHK7f/DbhiVIUyq+3QQ/PvUUdNbTnMzMzMzGaYvpXCiLgFuEXSLyLi3Oo2SR8F3jbKwpkN7J//nOoSmJmZmZnNSIPOKdyhx207tVkQs6Ece2xezMzMzMyslsnmFL4OOABYV1J1uOhS5NBSMzMzMzMzm8Emm1P4NeAs4MPkWoUdf4uIP42sVGZ1HXJI/j366KkshZmZmZnZjNN3+GhE3BcRN0fE3Ii4pcwxfBRwgKSr5k8RzczMzMzMbFQGWrxe0irAHuTyFBuRPYdzR1gus3rcQ2hmZmZm1kjfnkJJr5Z0DnAuuSzFq4A7IuK9EXHlZMEl7Sjpekk3SDqsx3ZJOqZsv0LSpoPsK+n1ZdvVkj426JM1MzMzMzOz8SbrKTwW+D9gr4i4GEBSDBJY0pyy/w7A7cBFks6IiGsqd9sJWK9ctgSOA7bst6+kZwO7AhtFxL8lrTTgc7UF2YEH5l9nIDUzMzMzq2WySuFjgZcA/yNpZeBUYOEBY28B3BARNwJIOoWszFUrhbsCJ0VEABdIWrYMVV2rz76vAz4SEf8GiIi7BiyPLcgWX3yqS2BmZmZmNiNNlmjmnog4LiKeAWwH3AfcJelaSR+aJPaqwG2V67eX2wa5T7991wf+n6RfSTpX0uaTlMNmg6OOyouZmZmZmdUy6OL1RMTtEXFURDwV2A349yS7qFeYAe/Tb99HAcsBWwFvAU6VNM/9Je0v6WJJF999992TFNXMzMzMzGx2mizRzKa9bo+I6yPivf3uQ/burV65vhrwhwHv02/f24HvRLoQeJhMgtNdxs9FxGYRsdmKK644QRFtgbH//nkxMzMzM7NaJusp/JKk5SQtP9EF+OIE+14ErCdpbUmLAHsCZ3Td5wxgn5KFdCvgvoi4Y5J9vwtsCyBpfWAR4J56T9sWOP/1X3kxMzMzM7NaJks0swzwa3oP5+zoOTYzIh6UdBDwI2AOcEJEXC3ptWX78cCZwM7ADcD9wMv77VtCnwCcIOkq4AFg35KoxmazD394qktgZmZmZjYj9a0URsRawwSPiDPJil/1tuMr/wdw4KD7ltsfAPYeplxmZmZmZmaWBk40YzatvfzleTEzMzMzs1omGz5qNjOsvvrk9zEzMzMzs3m4UmgLhve9b6pLYGZmZmY2Iw00fLRkB91b0rvK9TUkbTHaopmZmZmZmdmoDTqn8DPA04C55frfgGNHUiKzJvbeOy9mZmZmZlbLoMNHt4yITSVdChARfy7rB5pND49//FSXwMzMzMxsRhq0UvgfSXOAAJC0IvDwyEplVtc73znVJTAzMzMzm5EGHT56DHAasJKkDwLnAR8aWanMzMzMzMxsvhiopzAiTpb0a2A7QMBuEXHtSEtmVseee+bfU06Z2nKYmZmZmc0wA1UKJW0FXB0Rx5brS0naMiJ+NdLSmQ1qk02mugRmZmZmZjPSoHMKjwM2rVz/R4/bzKbOYYdNdQnMzMzMzGakQecUKiKicyUiHsYL35uZmZmZmc14g1YKb5T0BkkLl8vBwI2jLJhZLS96UV7MzMzMzKyWQSuFrwW2Bn4P3A5sCew/qkKZ1fa0p+XFzMzMzMxqGTT76F3AniMui1lzhx461SUwMzMzM5uRBs0+uiLwamCt6j4R8YrRFMvMzMzMzMzmh0GTxZwO/BL4KfDQ6Ipj1tDzn59/zzhjnk3bHnDrQCHO+cwabZbIzMzMzGxGGLRSuEREvG2kJTEbxnbbTXUJzMzMzMxmpEErhd+XtHNEnDnS0pg1dfDBU10CMzMzM7MZadDsoweTFcN/SfqrpL9J+usoC2ZmZmZmZmajN2j20aVGXRCzoey0U/4966yRP5TnKJqZmZnZgmTQ7KMCXgqsHRHvl7Q6sEpEXDjS0pkNapddproEZmZmZmYz0qBzCj8DPAxsC7wf+DtwLLD5iMplVs8BB0x1CRobtOcR3PtoZmZmZu0btFK4ZURsKulSgIj4s6RFRlguMzMzMzMzmw8GTTTzH0lzgIBHFrN/eGSlMqtr++3zYmZmZmZmtQzaU3gMcBqwkqQPAi8G3jGyUpnVtcceU10CMzMzM7MZadJKoaSFgJuAtwLbAQJ2i4hrR1w2s8G9+tVTXQIzMzMzsxlp0kphRDws6eMR8TTguvlQJjMzMzMzM5tPBp1T+GNJLypLU5hNP896Vl7MzMzMzKyWQSuFbwK+CTwg6a+S/ibpr5PtJGlHSddLukHSYT22S9IxZfsVkjatse+hkkLSCgM+B1uQ7bdfXszMzMzMrJaBEs1ExFJ1A5dspccCOwC3AxdJOiMirqncbSdgvXLZEjgO2HKyfSWtXrYNvsCbLdhcIRxn0LUPve6hmZmZmQ3UU1h69PaW9M5yfXVJW0yy2xbADRFxY0Q8AJwC7Np1n12BkyJdACwraZUB9v0EmfgmBim/zQL/+U9ezMzMzMyslkGHj34GeBqwV7n+d7Inr59Vgdsq128vtw1ynwn3lfR84PcRcfmAZbfZYIcd8mJmZmZmZrUMuk7hlhGxqaRLASLiz5IWmWSfXklpunv2JrpPz9slLQEcAfz3ZAWWtD+wP8Aaa3iI3ALvVa+a6hKYmZmZmc1Ig1YK/1Pm+QWApBWBhyfZ53Zg9cr11YA/DHifRSa4fV1gbeDykgh1NeASSVtExJ3VwBHxOeBzAJtttpmHmS7o9t57qktgZmZmZjYjDVopPAY4DVhJ0geBFwPvmGSfi4D1JK0N/B7Yk7Hhpx1nAAdJOoVMNHNfRNwh6e5e+0bE1cBKnZ0l3QxsFhH3DPg8bEF1//35d4klprYcCygnrjEzMzNbcPWtFEpaOyJuioiTJf0a2I4c2rlbRFzbb9+IeFDSQcCPgDnACRFxtaTXlu3HA2cCOwM3APcDL++37zBP1BZwO++cf3/+8ykthpmZmZnZTDNZT+G3gKdKOjsitgOuqxM8Is4kK37V246v/B/AgYPu2+M+a9Upjy3AXve6qS6B1eTeRzMzM7PpYbJK4UKS3g2sL+lN3Rsj4n9GUyyzmvbYY6pLYFPMlUwzMzOzZiZbkmJP4F9k5XGpHhez6eG++/JiZmZmZma1TNZTuGNEfFTSohHxvvlSIrMmdt01/3pOoZmZmZlZLZP1FL68/N1txOUwG84b3pAXMzMzMzOrZbKewmvLsg8rSrqicrvIPDEbjaxkZnW88IVTXQIzMzMzsxmpb6UwIuZKegy5NMTz50+RzBq4pyxVucIKU1sOMzMzM7MZZtLF6yPiTmDj+VAWs+Ze/OL86zmFZmZmZma1TLZ4/akRsbukK4GobsLDR206efObp7oEZmZmZmYz0mQ9hQeXv88bdUHMhrLLLlNdAjMzMzOzGWmyOYV3lL+3zJ/imDV055359zGPmdpymJmZmZnNMJMNH/0b44eNjhMRS7deIrMm9twz/3pOoZmZmZlZLZP1FC4FIOl9wJ3AV8j5hC8Flhp56cwGddhhU10CMzMzM7MZadLso8VzImLLyvXjJP0K+NgIymRW3447TnUJzMzMzMxmpEErhQ9JeilwCjmcdC7w0MhKZVbXbbfl39VXn9py2AJj2wNuHfi+53xmjRGWxMzMzGy0FhrwfnsBuwN/LJeXlNvMpoeXvSwvZmZmZmZWy0A9hRFxM7DraItiNoR3vGOqS2BmZmZmNiMNOnzUbHrbfvupLoGZmZmZ2YzkSqEtGG68Mf+us87UlsOsj0HnKXqOopmZmc1PrhTaguEVr8i/XqfQzMzMzKyWgSqFkg4GvgT8DfgC8BTgsIj48QjLZja49753qktgZmZmZjYjDdpT+IqI+KSk5wArAi8nK4muFNr08MxnTnUJzOY7D0c1MzOzNgxaKVT5uzPwpYi4XJL67WA2X11/ff59/OOnthxmM5wrmmZmZrPPoJXCX0v6MbA28HZJSwEPj65YZjW95jX513MKzczMzMxqGbRS+EpgE+DGiLhf0n+RQ0jNpocPfWiqS2BmZmZmNiMNunj9w5LWAvaWFMB5EXHaSEtmVsfWW091Ccysh1EMR/UQVzMzs3YtNMidJH0GeC1wJXAV8BpJx46yYGa1XHVVXszMzMzMrJZBh48+E9gwIgJA0pfJCqLZ9HDQQfnXcwrNzMzMzGoZtFJ4PbAGcEu5vjpwxUhKZNbEkUdOdQnMzMzMzGakvpVCSd8DAlgGuFbSheX6lsD5oy+e2YA233yqS2BmZmZmNiNN1lN4VJ9tMVlwSTsCnwTmAF+IiI90bVfZvjNwP7BfRFzSb19JRwK7AA8AvwNeHhF/mawstoC77LL8u8kmU1kKMzMzM7MZp2+lMCLO7XW7pG2AvYBfTLSvpDnAscAOwO3ARZLOiIhrKnfbCVivXLYEjgO2nGTfnwBvj4gHJX0UeDvwtkGerC3ADjkk/3pOoZnV5GymZmY22w06pxBJm5AVwd2Bm4BvT7LLFsANEXFj2f8UYFegWincFTipJLC5QNKyklYB1ppo34j4cWX/C4AXD/ocbAF29NFTXQIzM2DwSia4omlmZtPDZHMK1wf2BOYC9wLfABQRzx4g9qrAbZXrt5O9gZPdZ9UB9wV4RSmTzXYeNmpmCzD3ZpqZ2ShN1lN4HfBLYJeIuAFA0hsHjK0et3XPQ5zoPpPuK+kI4EHg5J4PLu0P7A+wxhr+kVzgXXRR/nXCGTMzMzOzWiarFL6I7Cn8maQfAqfQu8LWy+3k0hUdqwF/GPA+i/TbV9K+wPOA7TprJ3aLiM8BnwPYbLPNJk2KYzPcW96Sfz2n0MzMzMyslskSzZwGnCZpSWA34I3AypKOA07rmt/X7SJgPUlrA78nK5d7dd3nDOCgMmdwS+C+iLhD0t0T7Vuykr4NeGZE3F/r2dqC69OfnuoSmJmZmZnNSAMlmomIf5DDNE+WtDzwEuAwYMJKYckOehDwI3JZiRMi4mpJry3bjwfOJJejuIFckuLl/fYtoT8NLAr8JFe04IKIeG2tZ20Lng03nOoSmJmZmZnNSANnH+2IiD8Bny2Xye57Jlnxq952fOX/AA4cdN9y++NqFtlmg/PPz79bbz215TAzmwGcuMbMzKpqVwrNpqXDD8+/nlNoZjYlXNE0M5u5XCm0BcNnJ+24NjMzMzOzHlwptAXD4x8/1SUwM7MWuefRzGz+WWiqC2DWinPPzYuZmZmZmdXinkJbMLz73fnXcwrNzMzMzGpxpdAWDCecMNUlMDMzMzObkVwptAXDOutMdQnMzGwaG3SOInieopnNPq4U2oLhpz/Nv9tvP7XlMDOzWcPJcMxsQeFKoS0YPvCB/OtKoZmZzVCjqGS64mpmg3Cl0BYMX/nKVJfAzMxsgedKptmCyZVCWzCsvvpUl8DMzMxq8lxPs+nBlUJbMPzwh/l3xx2nthxmZmY2pdybaVafK4W2YPjIR/KvK4VmZmZmZrW4UmgLhlNOmeoSmJmZmZnNSK4U2oLhMY+Z6hKYmZmZmc1IC011Acxa8b3v5cXMzMzMzGpxT6EtGD7+8fy7yy5TWw4zMzNboDhDqs0GrhTaguFb35rqEpiZmZkNpO0Mqc64asNypdAWDCusMNUlMDMzMzObkTyn0BYM3/lOXszMzMzMrBb3FNqC4Zhj8u8LXzi15TAzMzMzm2FcKbQFw+mnT3UJzMzMzBYYnqc4u7hSaAuGZZaZ6hKYmZmZmc1InlNoC4ZvfCMvZmZmZmZWi3sKbcFw3HH5d489prYcZmZmZmYzjCuFtmA488ypLoGZmZmZ2YzkSqEtGJZYYqpLYGZmZmYTaDtxzaDx6sSczTyn0BYMX/1qXszMzMzMrBb3FNqC4QtfyL977z215TAzMzMzm2FGWimUtCPwSWAO8IWI+EjXdpXtOwP3A/tFxCX99pW0PPANYC3gZmD3iPjzKJ+HzQA/+clUl8DMzMzMZrDZvDbjyIaPSpoDHAvsBGwAzJW0QdfddgLWK5f9geMG2Pcw4OyIWA84u1y32W7hhfNiZmZmZma1jLKncAvghoi4EUDSKcCuwDWV++wKnBQRAVwgaVlJq5C9gBPtuyvwrLL/l4GfA28b4fOwmeDEE/PvfvtNZSnMzMzMzICZ1fM4ykrhqsBtleu3A1sOcJ9VJ9l35Yi4AyAi7pC0UpuFthnKlUIzMzMzW8CNqqKp7KRrn6SXAM+JiFeV6y8DtoiI11fu8wPgwxFxXrl+NvBWYJ2J9pX0l4hYthLjzxGxXI/H358ckgrweOD6AYq9AnBP7Sc7f2NO93ijiDkbyzgbn/MoYs7GMs7G5zyKmNM93ihiuoyzI94oYs7GMs7G5zyKmNM93ihiTmUZ14yIFXttGGVP4e3A6pXrqwF/GPA+i/TZ94+SVim9hKsAd/V68Ij4HPC5OgWWdHFEbFZnn/kdc7rHG0XM2VjG2ficRxFzNpZxNj7nUcSc7vFGEdNlnB3xRhFzNpZxNj7nUcSc7vFGEXO6lnGU6xReBKwnaW1JiwB7Amd03ecMYB+lrYD7ytDQfvueAexb/t8XOH2Ez8HMzMzMzGyBNrKewoh4UNJBwI/IZSVOiIirJb22bD8eOJNcjuIGckmKl/fbt4T+CHCqpFcCtwIvGdVzMDMzMzMzW9CNdJ3CiDiTrPhVbzu+8n8ABw66b7n9XmC7dkv6iFrDTaco5nSPN4qYs7GMs/E5jyLmbCzjbHzOo4g53eONIqbLODvijSLmbCzjbHzOo4g53eONIua0LOPIEs2YmZmZmZnZ9DfKOYVmZmZmZmY2zblSaGZm1oOklae6DGZmZvPDSOcUznaSFgN2iYhvthRvSeAFwNyIeG7NfZfvtz0i/tSgPD+OiP+uu1+feC/stz0ivjNE7LWBJwEBXBsRNzaMs3RE/HWCbWtExGArig7+eCtHxB8b7tvKcx4FSXOA5SLinnJ9EWA/4I0R8cQWH2cbYK+I6Dl3eYi4C0fEf9qM2aAMz4mIH02w7SVtHXemM0nPZuwzfk1E/KyFmMsALwL2Ap4IrFpz/037bY+ISxqUaUVgxYi4puv2JwF3RcTdDWLuHRFfLf9vExH/W9l2UER8um7MmUjSusBcYM+I2LClmKuXeEe2Ea8No/49aOs5S3o0mXLiH0PEWBP4S0TcV64/G9gNuAX4dEQ8UDPeEyLiuvL/ohHx78q2rSLigqZlne4krUTm/njkOAt8pul5SdvmR/mGOfeeIF7rx5wSd+jznVnbUyjp1Mr/H+3a9uMh4s6RtJOkk8gD0B7NS5kny5J2K+W9g0yyc/wku/Xya+Di8vcPlf87tzfRc/HLIezS5/K8JgElLV1eu7OBVwCvAn4q6ZuSlm4Q8ueV2Gd3bftukzJ2k7SMpFdI+inQ5ASy7efcHX9hSU8pB+Mm++8J/Am4QtK55Qf7RmAn4KUtlG8TSR+TdDPwAeC6YWOWuJK0raQvkGusDhPr2ZIOknRgef5NnCnpZ5J6VVrePkz5Otoop6QnVP5ftGvbVg1jrirpV8B7gHWAxwHvkXThBK/HZPEWl7SHpNOBq4D/IT87q/ffs6eLgS8DR5XLxyuXoxrEA/gUvY+3qwGfrBNI0lLl3zd1xa96RZ2YldgrSXqvpG+V48172+htlbSYpA0lPak0tg4bbxVJh0i6ELiazHI+d8iYK0h6naRfkL8TQz3vsiTX8yQ9V9I6Q8QZ2e9Bm89Z0gGSbiXPm26TdIukAxqGOxVYssTdBPgmma1+Y+AzDeJ9rfL//3VtaxIPyIbwfpemcXs8zuqS3tJgv23IJeIATgK+Wv7/VdnWGknbSDp2upSvxXPvTrzWjzklbrvnOxExqy7Ac8vfSyu3XdJ1n0sbxH0G+YG5Dfg2cCewxBDl3AE4Afg9+UHfBbi5pdeg9vObIM6NwAsnukz1e13KeCJ54rhQ5TYB7wJOGua1634dh3ldgcXJBoTTy2foL8CzquWewud8PPCk8v8yZEvcleWzObdBvKuAx5X/NwX+DbxgyPd5/fL8rgXOA14P3NLSZ2hL8sT7VuDv5PqoyzWMtSrwK+BcsuLxifL/hcCqdT+LwKvL5+UlbX0WR1DOS3r93+t6jZinAfv1uH0f4PQBY3wH+C/g5PIafrEcd+cANw3x2r2xfAZ/ALwMeHQLn8Gr+2y7qmass4Ent30sA7YhT+jfCzwf2LX8fzOwTcPn/SjgY8A9ZAPmpcDd5baFG8R7NXAO8BvyBGqjId/rpcpn7ofk7+HHgduHfK+XJis1N5bP6Gnl/28CSzeIdyLt/h6M4jm/g8w2v07ltnWA7wHvaBDvisr/RwEfK/8vVN02QJyvl/fj0sptl3bd59I6Zeva92Gy4feEcvlS5XLCkK/pCsDrgF8AvwOOahDjAuApPW7fBPjVMOWrxPlYOUb8DHj9VJePls+92z7mlJijO99pI8hMupQDxEtp8USF7DU4nzwBWKrcNuyb/jB5ErZ25bYbW3oNGp2I9Yhzb48D2VAHNPKEbIXK9UWA/cnhLk3i/bbJtkFeu2E/N5X92j4pbfs5X135/xDgu+X/xzT5Qezxul3X9LlWYnS+L4+r3DbU9wX4IPBb8iT6VWQFovH7UmIOXZnpfh3LD8RF5Xu3RK/XeIrLeWmv/3tdrxHz+ibbuu73eHIEwmXAFcChwOptfHZKjLWBw8nK9anAJkPE+s2wz7dy/yXJxphWj2WM5gTtE8AXKL+r5balydTrn2wQ74FynNisclvj9xr4Z4n3/xjL5j7scedE2q3Etf17MIrnfD2wWI/bF+/32e8T78rK/5cAz6lcr1Mp3IRs5G79d7/s+wLgFHJ0wTup/H7VjLNu+dtqhZ0ckl972yQxW6vQjKh8rZ57t33M6Spja+c7ncusm1MYEYcq5y+9U9JTyJajxcv/KpfFa4b9NjlefQ/goTIEKYYs6lOBPclhHjeSB445Q8Zs2y0R0WiYUS9lWOFngX9I+i35w/gV8oS36bBCtVO6R6wk6U0lbuf/zuM0HU67IfBn8iB5XUQ8JGmYz0/bz7k6/2IHssWaiLhTavRQ1dcN4NHV6xHxPw1ivoj8vvxM0g/J78uwr8P+5MnKccD3I+JfQ74vABtExAu6b4yIkyQd0SRgRPxG0tPIVshLJe0zZBmh3XLGBP/3uj6onsdCSQtNtG2eQkVcT76/3y9DXPcij7d3AUtJekxE3NmwfETETeW3YHGywXB9sgLaxG8l7Ry5fu8jJO1EnvzVKdc/gEskPUHSFeT3ZN3yP+V6k+GKS0fEpT0e77LKkNW6ngesH+Wsp8T7q6TXkcOkDq4Z77HAS4D/KcNaTwUWblg2yEr/nuQx4muSvjFErI5tImK/6g3l+b+v/C7W1fbvwSieMxHxrx63/VPSww3CnVMZ9rcc2VODpFUY/3s2WZkuAy6TdLykY8jXcrXyP+V67eHqlfinAacp56ztCnxc0n8BR0TEuTVCfaKU6XvkaI53AOdFREia5zhegyQtFxF/7rpxeZpPP7sO+CWZb+OGEu+N06h8bZ97t33MgdGc7wCzeJ1CST/rtz0ias2fUZ4dP5scI7wz2Zr5SuDMiPh703KW2NuUuC8iTypOi4hai1R2nYS/iRwS9ogmJ+KSLo2Ip9Tdr0+8q4DdIuIGZbKG/yMn4p42RMwvk0Mn3l89sZD0TvJk42U147273/aIeG/DcnZOSvcA7gKeADy5yUnpCJ7zz8gWx9+TQzyeUCqEjyKHrj2hb4B5443kNSyxlyQbaOYC25Jzu06LiNrzhJXJcP67EutnwPZkb9KDDct3Q0Q8rsftC5Et4vNs6xNrnu+fpGeRvfcrRkTTE/G2y3kXYz9ae5T/Kdd3j4ja85AkfQJ4NHBIqeR03vtPAP+KiDfUjVmJvRn5XXwx2cq+dc391yF/sHclRwCcQmlUGKJM6wPfJ0ek/LrcvBnwNOB5EfGbBjHX7Lc9Im6pGe9aYOsJTtDOr3ucKPv+JiLWr7ttwNirke/TXGAJ8jhxeMNY65Q4ewLrAe8u8Zq8Lz2/e2XbbyNivZrxWv09qOzf5nM+G/hQRJzddfu2wDsbno/tAawCnBoRvy+3P4U8Ntb6PZC0b7/tEfHlOvF6xJ8D7Ei+lhsCh8UEScT6xFiNrHzsSY4G+BrwDeAnEdFoTqqk/cnhj4cylt/gqcBHgS9FRO05dqWSuiewNdmjeQrwhYhYu+XynRARn60bsyv+0OfeXfFaO+aUeK2d7zwSc7ZWCkdJ0sJk0ow9gf+OiBVairsQ2VOzZ0S8vOa+rZ+IS3pSRFxdd78+8S6JiE0r169rciLRFXNpcljmpuSXOoCnkHNTXhURf2kYd4UomTPb1sJJaavPuZyQHkMOFz06Ik4stz+H/Hy/uU68+aWcjL6E/L40TebSibUY2WsxF3g6cHZE7NUgTmuVGUm7RcR3e9y+HPCaiPhI3fKNqJytn1CVY+yHyYy1t5Cf8TXJH8XDo2Z2wQkeQ8AzarbYU3o2riDnB/+Vrt7Qhj3hnSQ9e5EnjZDJCr42TGWzK/5/kXPjb42IX092/x77t36CJum7wHci4qSu2/cmGxSeXzfmBI+zPjk/unGDVCXWk8n3afeIWLfB/m036vX7PXhllAydwyjPeS6wR8Pn/CTy+3Ie2egRwObkPNVd2zrPkPR08n1uLRu1pDXrNqBU9u10JGwB/BQ4JSKaJv6rxm2zwv484K2Mz+55ZER8b8gytlKhabt86pFVf5hz7z6P83jy+/K+luJ1znf2iIhtG8dxpXA8STsAb42IHVqKt3hE/LPBfq2mNdcIUoxLuonxJzyqXI+6Pw6Sbmd8D+a4Hs2mJ1Ml9rrABqWMV0fE7xrGeR45d+s/5Lju3SPi/KblmuSxFinxvzrpnXvv38pznuQxNo+Iiya/57h9jum6KchEEj+LiPNaK9yIKIfCvXCqKzMawTIzoyhnibdi2f+Gpg0xE8RdnMw8qhL7ftVcxkXSp5h3iOs9wDlRWaahRrz3dMUbp2nFQ72XcdkXeFM0WMZF0vfJHomrlMPqLiHnNq0LfC4ijm4Qs+0TtFXJZCv/ZHxlYXHyO1grC7BGuOxRW9quxKkslTTq3wNJ/02eO23fcP/FyMr0kyhlBE4ettFDmX10L2B34Cbg203OhZRD9FcFfhERd0naCDgM+H8R0SRLcbUB6Tzyfe5uQKo14kHSh7p7noatsE/wOG0vt9aoQjOi89pLo90RcF8H3hcR1/bY9pUmPfUj/e2frZXCMizheHK873eBD5EpbQV8sM6Pg6Qr6X8SsFGD8lWHtz6VsSFDJWS9loDuXrg2lJblqoXIA++h5OTrF9WMN4rezLYr11eQFbXrJG1JZjV7Zt1ydcVcmlxnZ1XgDOAn5fqhwOURsesw8SuP83jg0Ih49ZBxNmBsCMR9EbFZzf179RwtT352vtHwZPRvjH0HO2Prg8xeuEhE1J4/rRxyfV9EfLHr9tcDc5qUsxKjjcpMp1Gm11yCiIZDhkZQzleRx9ffkclX9o+IM4YtW9djLENlbcGIGHiezyg+j21TZb41mfzoPYzNt35/3eNYiXl1RDyp/H84OSx8n9Lo8b9Nfrf6PNaSMdy6c9tSqSxE1zDDGnG+1GdzRM058l3HnXGbSrzaSz60XYlr+7d/knOnD8QQ0z16PNYcsnfm5Jr7rc/Yb9S95DDKQyOi75DpPvGOZCwp1ePIodwHkM/9s00rrpL2o/+5Y62Gx1Gc51ViV6dTPAf4ZUS8uEGc1io0IzqvvZE89+qpbsORpLuB+8lzxWO7tjUq/yh/+2dzpfBSMnX4/5FDPU8ix67XWvOpxOocaESmId+5ur3p0IJqWYdtuRjxwWIhMpHCW8iD5oeia5HlqaL+c0eHrly38boqk1H8mfwsbkdOjF8EODhyonvdeBuRWXY7P9qfItdS2hL4eER8okHMNckfg7nAg2Svz2YRcXPdWH0eY3Fy7tHQrXTlxPYA4DXkkJTaQ1yVc1w37e4VUw7ju6iNk+YhKzNrDntsqfFYw5TzKuDZEXF3GdZ0ckQ8rWE5HumZLp+X55cybUpm3tuNbMlvkpyi+7EafR579ISPU7f1v8QcxXzryyJik/L/2cDnI+KU7m01Y65KzuO6IiIeUK5legiZyfaxTcva9RgjWfh5WG31MIygEtd2z0dr506VmNWG0dPJoZQHUs4p6jaMlh64X5I9q51kJjc2PVmWdA35W/Av5fD8PwAbRUSTxD8jI+lycimrnklHBq1wSdorIr5W/n8GeYx9LpnEZhty6ZD7G5axtQrNiCqF95KfwYnKV7fh6FLye/JlcoTZfjE22qPV72YbZl320YqIiJ+X/78r6e6mB7XqiZmkf4/gRK2NmvtGkv7a4/ZhWjQXJhfDfSM5/GHXpq2ZJV7rwwpjyLlkPXRnzhx3PZoNcV0nIp4MoFwY/R5gjYj4W8Myfp7MDPd/5OT1S8hJ5y9t0qIp6XxyfcJTgBdHxG8l3dRmhRAeyTQ3VAxJy5InofuQz3nziLi3eZHmHSYZEf9WzYIOWpmpWb7Tyv6tGVE5H4iIuwEi4kZ1LWBf0yuBiyR9jUyJ/2Pg02R2wRsqx/ShDfF5rD0fbwAPdE5uI+KS8v0btkfmttLrfTv5/v4QHnnfa2fHk3QIcARwA7CopE+Sw/9PIke71Ik1LqOgcnjrHuRncSNyaHOjhZ/VexjufsAbo8Ew3Iq2Wtjbzha6ar+GigaNFK2dO1V8hbGG0VeTQ5AXIc8pLmsQr+3sjP/s/HZGxJ8lXd9GhVDS9+jfU1h3zuwTyONPzwoNg2cV3o7MLHs7uT7vccBbIuJv5djTqEIIEA0SyvTR+nktLWfVL+W4E3hOOVe8WNLrIuKsFh8DaGc02GyuFC6r8fMLVL1et4t4BrhyBC0SN5G9RkeTB46NJW3c2djgNex1MrU8cKSkpsMKn9Fnc0TEL2uG/Dx5cjzR9Sb+UynQQ+Wg27RCCLBolGQwwPWSDiXnDj3UMN7dwGrAyuSyG7+lvRMgAJSZTF9GnqA22X8F4M3kieMJ5HppbSRQmGeopDKtdF2jqMy0ffIIoynnal0npeOu1zwp7STtehLtLuMyzjCfxxgyE+EERrGMyyuB95EZdfeIsbmeW5HzpuvaH3h8RPxJ0hpk5fAZEXFBg1jvA14v6dVk5W81MpX7q8h1MpvOyxzFskdta7sS15mP2ZZRnDu12jAa45d62I1suF5Z0nE0y864rqTOkHcBa1WuN6m8dRzVcL+JXNPSed6S5W/ry60pE9QtFRHf6rp9L+DuiPhJjXCjOK8dxe8qkMfpMirjq5J2Jhs+aptsNNgwZZzNlcJzgV0muB7k5PaBaPy8teqahxms2XyPavKD7pOqRkOQRuCnZBk3LpeqWq8hTHwyJel4MhX70fWLyFt6PRRZ3tWouf5M05ORSWxcae0S+Rn6K81buxbr+gz+nWxRE9T/PEbErpXhg++V9DjyxGCLiLiwZtmq83CqB9/7ye/ga+rGK24hK69fKrFeWe3laXjSfCTwA0lvZnw2xY9R/8d8FJWZtk8eYTTl7P4ODnOC+lWAiNhY49cWvJuGawtq3nlhYojPY2mgOJB8/U4gP0f/j5xT+eZOj19N/RqjGr0vEXEX8NpS5kerzPuLiJ+Ry6/U9a/O8LTIOXG/aVghhLE1X48le4/2ipKVccjK/zuAp0ZLw3C7KkfdlaWmFaS2K3H3ttxQ0dq5U0XbDaOdWP8ATgZOVs5lezG5zmLdSmH38NW2KnM3RcStLcVq03sBIuLgMgKgkyX1SGBpSbvTfLm19zL+89NxDjn6pU6lcBTGJX7RkFmZ6apkRsTlkjYnK29NRya0OhqsatbOKeynV+/AJPdvdd5aidlqGndJh0fEh+qWY7poa+y1MiX1EeS8vQ9Gzax4/U7CYXpU1iX9nIlPFBt9Hrvir0y2HO5JrtnXKPNamzS6jI87kRnmOnOXrgI+UnfohyrLR6ilNSkl3QK8a6LtTU4E2yynpJVKxWPkNOQyLi2X5cdkFs+lyGFYXyIXlf5/5I/2s1p+vNoZgCv7vg54O9kzIOBvwEcj4jMNYnXWo+zYs3q9zrFR0hER8cFSwX4JeULaWfh5v6bHHLW87JFaTlxTYrY9p/CCiNiqrXijIOkhMolS5wR6cbJhZpjpLRMNE37TkO/5imSh7m4aoxLrkfda0rejZnK+HvH2q4wS6t72qKi5vm73ubByytCO5Pex0XJrkq6ICebj99s2wf1bP69Vy1mZJS0UE8xxl7RKRNzRoIzj5nxLug1Ya4jRYGOxXSlMGiKRwojKM09q4SHjnRoRu5f/PxoRb6tsm2ddlhpxNyR7AqopyI+KiCtbKHZ1GNcLI6JX69KgcbYD3lnK+KGaQxSqcVqrrHcO4BpheuFRU8NkJ+UH+qWM/9x8LSL+3XIRp7VSmZlLnvjWqsy0ffI4yWPVLqdyLs97yZb5NufNdD/OozNM/KP0htdaW1CZROkvUYYbK9cO243sff501F9+4/LSkylyfsoalW2NErj0eIyhMgCXGO8gF5A+KCJuLLetA3wS+FVEfKBmvNbXo+yKP/TCzxrhskdtabsSpxzKO6G6PVUaP4y5V7zp8Bq2mq23fJffBbyerKguRE6d+VQMsc5ctbG7jYZvSedFxNPL/+OWO2jyeyHpTuBK4OvkUh73VbY1XW7tN8AG3RXUUuG8JiLWqxGrezmhcZo00qvlrMySjo6IQ8r/B0dl/q2kEyNivwZlvI48DnYaUU4m6y6NRoNVzebho6ilRApqf94aZGtMa5VCMo1yxw7A2yrXV6QBSbuSwyg+THaFixxe9x1Jh0bE6TXj9Urv/U+GGFYo6blkz+B9wBHRYN2xLt8gx8OPayVUZtnrNeG5n+Ul7UMZqjGBOpPDO2VpdS0uTTIZnvwO1Ym3Abn0xv8yNin+WcARkp4fDTLXjujHofWYPWJcTE48P5QcolJHz8qKGqZx76dhOXcml6Boe94MAJIOIHtxH12uN+3lOhV4AXCfck2zb5LHtI3JeRqvqhnvIcgDv6R7urY1zoqq9jMAvwzYOCrDjSITAe0OXA7UqhSS8wnb/M0aJ3JNwqOAo1QWm28QptVhuOX4PZGIiK/UjQns3q8iV7cSR2ZE7x6uH+Tv/krUnELB8HPoJ6RcU6/Ti3dNNF+0vtVhwmTysqeTictuKmVdBzhO0hujQUbvont91GEtWfn/SV3bmsyVW5Wcc7wn8GFJ/0dWEM9oUiEsvgN8XrnG4D8AlHM/j6H+0OOLG5ahn/9U/t+OPEYQmWSnyfG7+nu5L9no1tE0g/kdjG/curNyPYDGo8FmbaVQ0snkm9VGIoVW560Vc5Spj4dKLTygpgej9wE7dJ2UXC7pHDKlb61KYUSM4sfme2SyiHuBt6krm2CDXopjyCx93QevHcgfjdcNGqjSonpSzTJMpl+PapM5H52TepEHyLonyd0+Bbyuu7dW0vbkHKImGWNH8ePQekxJHwNujIjjuzYdAjyGbAAZ1HMkvZ3x61seRK6xdBnZejhl5SxDZn5XLq2q9HI9q7uXS9LyNXu5Fo+IP5T/9wZOiIiPK5fauaxB8dZRJqFQ5X/K9UaZ9zSiDMDRY/5JZNbVJic/bTdkTtbAVXs0SvQZRq6c51NXr31EHoNXJXun6mq1EhclgcsjhZPWIhuFtyfX2aul32vYVBmpdTqwBtkgIeDJkm4lM5DWbXBtO1vvPuS5ziONPKUBZW/yHLJppbCTT6CaSwCaD5vtdy5X+zyvDEf8EfCjMrpnJ7KC+ElJZ0dEk+RM7yAbnG5RToEQsDrwRXI0V53yjSKpV6tZmRn/PW4liU20n1X/EbO2UkjOEWolkUL3sEaNzVu7gzxJa6Kt1MIdSyiTjyzE+GQ4IsfvN7Fwr5OSiLi5DAWorQwXfai0sq9OZlP6XURc2rCMbX95nh4R+3ffGBEnl6EGjbQ5nDIiXt60HBPEe6QCIOnvdYbmTWDV7gpheZyflt65JlrvpRjRD87zGJufWPVJ4ApJ5wKLRJnXN4mTGEvj/iqycWqYNO4jKaekK5m3RfweMpnJUb0qJgNos5ereozdlpxjR0Q83N2INKBqUoruXtKmvaajyAB8u6TtomsheOXi5LXnuTCahsxvkRXzyzrFq4akWVKTR3QPwwVqDcONiNdXYok8hr8NuAD4YJMytV2Jq8RZjzwv6WQofENE/Kf/Xj3jjGIqyvvJRrhtS2MSpVHmI+Tr+Po++/bSdrbehasVwkqcu5ue65T9m3QY9LOspBeQ53nLVhpVRDYqNRa57ug15DnzU4ENGsZ5EDhM0nsZG8F2Q9OeR+Ww9YOBx5ebrgWOiYimje1tZ2VeqBwXF6r83zmONX7/y+i0Axl/znhsDDmPf9ZWCmPe7HV30TB7XYdamrdWtJVauKPa3Vztau5cb+I/ktboHs5ShjnVmtBc9ns18FHg75LeT57kXgI8RdIJEfHRujE7FRhJi5EHoCArmU0zNPU7S1yoUcD+wyl3rTuEZhRj2CvaGOKykKRFuyu85T1qekwaRS/F08lU6SeV698il0gB+EBEnNMgbESPSeelArIsOfznw2Sa6cm0vb7lqMr5vB63LU8OpfkUuS5ZkwK21ct1jqRTyWPkcuSoEZRJBmrNJyxlGLbRpFfMVjMAF28ATpd0HnncCbLnaxvmzbY4iLYbMiGf7x7kMKvTga9Hs+ytj2h7GG5pyNyPXBLnV2RP7vXDlLHEbasSt2GJ8yQyc/IrY7iEFNU5X61MRSFPwDeqHnPKseZwGvQI0/7SUf2OA7WPESN0LmPTObqzxNZdXxZ4ZE7qHuT3ZUlytMKuEXFtw3i9ph9s3mmAi4iBy1mGbx9Czg2+hDz2bEouY0aTimFUsjJ33d40K/MyjD8uNp7v1yFpGzLb6Ilk43DneV8o6aUxzDSpiPAlk+1sRlaUbgXOr7nvc8klE84CtmmpPJdO9WsyQBl3A35D/iA+mexZeDlwPbBbg3hXkydla5ATxFcoty8BXN2wjI8ifwjvIb+Yl5Kt7h8jW//qxjsX2KLH7ZsDv2hYxrPJoSndt28P/KxBvEt6/d/r+oDxlq9cLi/v0SO3NYj3DuD7ZLaszm1rkRXjdzV8Decp1zBlrLwvG1SuX0m2kD4D+GHDmBcB6/W4fT3g4vL/m+q+z03f2/lRzkke59Ih3pvtety+bd3vDPmDuie5ltmqldufAjynQdmuBK6oXC4v5X0HsFhL789KZO/J+cBtQ8RZDHgFWen4H7KVvFEZm76XA8ZekmzAPR04D3hmwzjnk78z7+x8vsllAZqW60DyN/A4YM2WnuuG5LytK8jhzHOGjPcQcDM5NP+Y7kuDeK3+vpT9LmuybX5dymv41x6XvwH/GSLu3ypx/la5fj/w4DR43ueTCbc+TjactBHzez0uZ5THeahmrAuonEdUbl8LuGCI8p0x0WWK34/NK8/7KT22b0ImCWv8GLO2p7BbjCVSOAzYvebubc9bgxyzPVFK5TdGRNP1TcaRtAPw1ojYoe6+EfFdSTeRLaSdrFxXA7tHxOUNivNARPwZ+LOkGzrPOyLul1S7NU7SY8nexqWAtaP0oEhampKwgBx2UMdbgFMlncjYWlKbkXMO9qxbxqLt4ZRtj2Hv9CL0aumq3QMQER+QdBDwC0lLlLh/J4cSNh0+OopeiqVjfNKb30ZZp0jShxvEg8xgd5akDzD+8/N2ssWTGHxoU3U+Cgy/vuWoytlPo951Wuzlivw1PaXH7U2HrI+kZ7QqsjX7U8CnSs9X0zj/ItdSnO7+RQ7v/CvZaLhYwzhtD8P9FLlcy9OB71V++zvfwSaJJC4HbiPnFm4BbKHxa67WTXBVe1mMSYxiKkr32rodAhatG0wtLx0V7Q/z7MQd13upzHB5AJlYr/YcSM2bGbYzVP+8KAlyano72dgdqqxl2iDOWIHanW61dEw8hanp799IEqN1k/R44NCIqPN7cLCkb5LPe57fp4i4rHyGGpu1lcLygTmQ8UkaDiSTNFxOWSR5QKOY9Pkv4E/APyR1p1SuPbm3zBM5HngsOeTrQ4x1Ozea+wBQKn/9MrDVsXjlx2aRrh+bWicBkr5Mnuw8D1i/nPh1yvxX5fpc11GzUhgRF0rakjxw71duvhrYMpqP5W57OGWrY9gjolFyjElifhr4dOcAFsMPeWx7uDXAstUrEVFNerFyk4ARcZak3cjGhc48mauBF0XNZVxGdaJSYrdWTmX2v27Lkb0gjYY0RcTVZVjcXuSwOJVYr4khF+8dVvReouUW4FJJtSuakvaNPvNbJ3i8QeL2yvbc8W8yQdAR0TXnsI9PTn6XepTLg8wlK0c/BT5ZGnAbifaH4bZ+bCR7a9sYpg9MPDe6/L40WeZpFFNRuuN0b6urySLjU6YMyT+EPJf6GtkjdG+DUL0qBGuRU1HeExHzNH71ExHnSnqdMqHZklnUxlmex2lpulW/eYhNs6NuDnwjIm5ruP84kjYiK5qdc+9PkVmtO8PCBxYReyuXMJOk5UonSvWxlqd5Q2vGqJwrzyqSTmcsScN25EnKIsDBUTNJg6SlY4LsWL3m3A0Y8ypyCGYrKZXLycgbS5ydyArhO6My36xBzDP6ba/bQ6r+i64TNTIuSfoR2dr244hYf4L7/GaibfOTMpPiVuR6YTeX29Yih/dcHDXXQZJ0M5n6fqKED7VOZNT+OleLkXMU/kQOI30LOSTzd+QaUvNM6B8g5qVtVwqVS3EcHxE/6Lr9eWT21Oe2+Firk9/vIxvs+2zGJptfHc0yKI+MpO55GEGOqvg5uRhw7XlSM5XKGoY19/kwmSH1kNGUqudjziGHMZ4cEb0SDvXa50tMfPyOiHhlg3I8TA6jPK/EHhe/Qa9Zd/yVyWPRnsDqEbF6zf2fEBHXlf/HNexJ2ioiLhimfD0er/YC5F37zwH+m6xoPwf4ZUS8uK3yzQRNXsNKA0p3oqNHkQm3GnWwSFqBHGm1B9mI/amorAXYllJZ+GnUX6fwCHL0RStrmZb9q8uEfSCGmP8m6X6g1xxjkXPul+yxbbKYnwBeDNxEDuP+ZpNzkkq8X5HDy/+PzH3wVrLi/86mDZiS9idHnBzK2Mitp5I5OU6IiM82Lu8srhReGWNJGuYwRJIGVRYFVabp3a7XtqYxy/XrIuIJ/fapGe93EbFu03glxt3kMJevkxPsx1VCYgQJFwZVeoI3JL+A34muCcfKVNK7N6i4/oz+Jz7bTbBtsrgHlbIuUW76B8MNp2yNxrJHdv8grgisVLfHSpnU4z9ky+NywFXkEOynA5tERK/hd5PF3C8iTuy6bTlyUfJGB7nSi/ADcl5F9cC7NfC8iPhNk7iV+CuQi8HPJUcsnBYRh9bYf1Uy++K/GBs6uyk5hOsFEfH7IcrW+gl+m0bQy9VaMqpJekb/HpWMlTViPi0i/q9JeYYh6TWDnmBIelGPm9cge0DmRMRqDR5/337b+/WgNnisNev2unb99nf/xjb97W91AfKy3zPIXvXnAheSJ/rrRMT9dWP1eYzGU1EkvTUiPlb+f0lEfLOy7UNRM7P0KF7DrvjjhnlGxJsbxvkHOaT5S+R8wnGinSH6nccauOG0VIzeSv42b9x9LFQuz3B5k0b10tBzOzkib55jeJ1zMk0ydH6IURQiG6r3JKcjXE6e555Wt44g6bKI2KRy/TZyHuQwyZ46jdNvZXz20SMj4nvDxJ21w0epLFAZuRzFTU0qhEX1ZHn5PtvqaDul8rIav+aTqtej5oLmxWPI7GNzyR+cH5DZ4RotOKsWF10vPbfnSzoQ+I6kVzB+7tHi5ILVdfU6ad+K/HI2Gj4q6YXR4nDK0rPwiegxnFVdKcQHEe2nSN8gIjZUZu27PSKeWW7/oaQmc1EB1ui02ktalFxbaGPgQUl7RcRP6wYsvfQbMbZUCOQQxdcOUWFYivzc7QWsT84bWafJCTO5vupxPSrD+5DDU2rNrevy/R63PXKCXzeYWp4fHX3WNK32ctF7WY3u+z+K/By/ghzmuRCwWqkYH9GgJ7N7SNC4ntGasTLAFFQIy+MO3OIcEd/u/F96Ew4nT6w+Qq5B1uTxH6n0SXp03tR8TtNko1sYy9w4cMgJ/u91fVCtLkAu6XYygd5xwFsiF+G+qWmFUKOZirInmfwNch7bNyvbmmSWbnsR99yxvWGeHUcyVjEaxTrNwCPv2Z8nveOYH5O9l9Hrty6ar2UKtDrdavF+PfXk8by20pB8LnBuabDfnjyOHc9Yw/2guufL/h3YqFQ8iYja2UglHVTOGXv9Tg9lNlcKO0kagHELhzZJ0hAT/N/r+qD6pVRuErM7PXH1etBgvafS0vFD8mR+UbJy+HNJ72vYwzWKNan+GBFbloNiZ+7RWXV6EKqiJBoBkPRMckz8omRF4awmMcmshN8p8dtYTuAAYC9JB0TX0EfmTSE+MLWUIp2SwjsiHpT0h65tTVvP9iDXuoJM6AHZk7k+8GVyPlJt5UemzUQcd5Et9e8gJ/+Hcl2pJjaIiHn2jYiTyrCfxto8wZe0J/BZWpofPZlyXLpcgydpOpIWk1HFCBcWnu4kPZE8RjyFfF1fG0MMdywxX0dWFJYs1/9O8zlNT6PP6JYGRvHb32+/JjG/TWYK3wN4SDl1ZpghYh8H9mdsKsoFDDkVhfYr162+hpp3mOdTooVhnhHxnmFjVGneNWEhOyr+QI3cD51zGUkvV7trmUJmKZ5wulXNWF8jR8dAfh6rPcCf6bpem6Qnkw0We5ANe02Wveo177ZzPciM2XW9gmwUbt2srRTWHfI2iU6vnhjfwycartsTEe+daJukzRvEe3mfeI2SZpR9FyWHpMwlJzQfQ/MFhVtfk4o8Ad80ck25JuvKzUPSc8jK4L+AD0auXzOd3Egu7n2ypJ3IDFed1r4mLc1tr3O1mjI7nCr/d8q2asOYD5TWPci5MqeUMl5beoJq6zNEcZjsnoeTPzLHAV+T9I0mZSt6HsOUiz4PfXxr8QT/HcBTo6X50YOq0cvVajIqaL9ntMTYl3YXae7EXZuxIUjXRpk71CDON8kMtUeR89cfApbW2PpjtRevV8633hp4VnTNaZK0fNSf09Tq6BZGcyxbVi0uQB4RB0s6hOyhmUt+l5eWtDtwZkT8vX7IR+Ytf1fS3UNWCKH9ynWrryHZ49QZ5nk/8EqNzwjbeJhn+Y1+O7kYfGcY4Ecj4swG4bqnXgRw7xC9662uZVqGnf6cUllT13Qrsue5TkWu9Z760vi9J/ldeYjMTP3fTY+LEfGsJvtNlVk7p7AXSUuSLWp7RY0kEpLe3W97vwpejcfYgLEP6n0RsdmQ8ToZ2PYCnhgRtX/AlBk+NyTXZzwlIq4apkyVuEuSB5w9gP8ih3A1mp+olhOQSLqIrOgfSZ7cjtNwKEC/ydIRNdOaq8yZUM6R+hjZErVXRFzR5PWQ9BBjKdLnqQxGzWQPGsE8IUkXAK8C/kiuk/nUKCm4NeR83FEoJ7Zzye/0esC7yfkKA89TVM77eDRwSOdHv3x3PgH8q+770hW7eoJ/Kl3ve50TfLU8P7pt6pNwqt+2PvEe6Rkllzx4D2M9o+9veIzYh6xovYmuRZrJjJy1K4alN/QL5Pt8WYm5MXny98qJWvP7xLuZsRP4zt/OiVlERO1lYSRdT8tzmioxOqNbjgQajW4Z0bHsS5PEnLCBd8D4C5NDMueSJ7sr1Nz/RsZPoziqej0aTEUpvzH/gEeWtegMbRW5bubCNeO1+hpKeg/951jXSgZXiftqcl7iW4FOVt3NyBEZX4iIRsPNK/FXZayB8A9NGvXKeUQ1y/PVZAKqWlMoJH2Q7J1/T4zNwx13PlL3/ESjmdN7IzmS4JSomWl7gnitzpct+z3I2Hdk3CaaN1hngNleKSwtuDuTH/odyaEW34kGkzUlrRBDZCnqEW9N8sA9F3gQWJNcQPTmhvEWJ+dM7EWeUCxFVoJ/ERG1x4crx5R3WqCqH6ShPpillX1H8oR5Q+CwiPhRw1i3M3Gq69otfOqfITUiovZQAElXk5/BiYI2Tn5Qru9EjoU/Bti7QaVwP/pnhW0l2UP58dmletCsse9WwIlkhf3oiHh/uX1n4GURMbeNMpaYywIHRkTjpVy64j2Z/I7vETWSP5WTuw+TPVC3kO/RmuRw2cMjovbanpXYN9PSCX6P7+CbqteHbGUfupdL0ndpNxlVq5mjS8wLSoybu25fizx52apBzBPJRc3f1zn+K7s/3gk8LiLaWmqoMUnXR8TjJ9jWqHGhx+iWM8iMfY0SM0lakfze3RARf2kSY5RK+VaM8WuudkaA/DXqZ4/uV+GKiGh7XcRpTdLmEXFRw32vAZ7e3cgm6b/I6QW1RhUol45YuFNJlXQrmeVzYeDLEVFrfV1JP46I/66zT59Y55AdEWe3VZGTdBfZkyeyE6Gz5IbIY3fjUXBtGVHFtdXOjqpZO3xUmSmrk5b5Z2RL7hZNWuGUWYC+BPynVJR2j4jzhyzf+eQwh1OAF0fEb5WTw29uGO9kck7Qj8mxyOeQP2I/b1rGiBhqPZRuanlNqmIO2Zsy7NwRYGRDAR6oW/GbRHcW2LMkbUbOhai9mHJ0JTJpk3qkSGd8koGBRKZ+n+cEMXIITpNhOCiXiXgnYwkVvkbOW+wkGWhFaY28spys1tnvP8Chkt5JZs0U+Z0eOqNgRKw1bIyKtudHT9jLJalJL1fbyageiDLsPSIuKcftYYfKjmKR5m0iYr+ueAG8Tzn3sxZJe0fEV8v/20Ql1bzGEiPUdbtanNPUNbrlvTHk6BZJryITrfwOWFvS/hExWTKbyWJ2L0A+ToNGlE+Rw9W7rUoOZ9+rTrBheyp7KQ2CryWPY1eQlfRhlt5o+zXsjj9u5BZ5HGoUqrtCCBAR90qNTlleAvy/yvV7I+Ip5Xf2XLIRsY5G058m8DLyuNrmdKu3VP7vPlcc9tyxLaNIRjUys7ZSCPyIPAF9eowNM2s6Lv5DwP+LzHy4JTlk75mT7DOZu4HVyEWyVySHIQ3TrbshmX3qWuC6yIyrQ3UTK9e+qQqGWAIAOJuxNakWBfYpw6YyeLPhcHc0HdpRJemL5CLeO/a7X5OhM8A86/RIWpcyvDAGXCesYp4FiSPibmAXSf+vx/37Uq7X16+nsG7GPtQ7RfraTSs0ko6Oso6bpIOjMsdF0ondJ78DOon8If02+b5fQA6deXJENF2kuZ+XkMkMaomIfwJDD3OpavMEP1qeH10cQ8692bNHL9enqZdU4fdAa8moaD9zNIxmkea2T0jeBHy1/P8pxs8NapoYodU5TeSJ6T/IBFRvqJx4Nx3dcgjwpIi4Wzkk/GSy53EYR5ENHWeRy6sM+z49OXpMv4iIH0mqtXh2h9qfM/tlMiP8L8lRM0+iwVzein6ZPJs2RLU6cqv4q6SNI2Jc1m1JG9NjiYpBxPj5g58stz1URorVtYz6ZIWvc75TjrO/l1RtFOxuMPxCncK1NUppxEaRjKp2w/mgZnOl8KlkS89PyxjiU2ienOHBKGlxI+JXKssKDCMidtXYvL/3KtdMW1bSFhFxYYN4G0t6Anki/tPS7b6UpMcMcYLb+ZGu/mgtJeky4FUNDpatt0DS3onPcWTFYJ4KV0XTLK4HAUhahRwCsRfZo/dh8geobrzbq9fLyfKzS9xdyIaGOo6qW4Z+1HKK9OIZlf/3pfwYFrV7R4vlYyw73I8k/ZFMQf7vPvsMYzq1Go7iBB9orZW91V6u4m4yMxzk3NSm+vWMNvVESVf0uF1A7bl6xf9Kehc5z/GRk5PS89xk0fXWW8Qj4uoyzLE6p+kXwGuiwbIwbY9uIXuF7y6xb6zb2z+BTcnvx3PJ39ivk0Pump5A9puPV2uuHjBuzqzayya8QYytG/1FsqGwsbYbotoeuVXxZuCMMiS32uixL7muaV2PlrRwGUXyyCif8rlsMqJgGTJ5Ta/vb9OM8Pc2HDUwD83H9XTLyKE9I+LImrt2VjqornJAub5Yw+KsorGkVvNo2IECzOJKYURcClwKvE3SNuQJyiKSziITPtSZ4NvdMjzuetOhCpEpj08ATlBmCN0DOFrS6hGxeoN41wHvAt5VhhPOBS6UdHtEbN0g3tq9bi8tS8czSa9aj3g9W33K0JJ+lbF+mqT7nUcZxnoxzTOrTkg52Xwu2TN8Kpkw5fR+P2wDxt2SPJl6AZmW+kDGD7cYSK9W5spjbNOgaG2nSIf+J6TNg0rLVeLdCSyhTObSNJtid+/6I5uYXpXCVk/wR9DK3uZ7vAyZ7Xh1cqSCgCcr5+PsWnMoaiuJxXpolLF0Eq8nlxe5oTTkBVkhuYQ8BtU1ihZxSuVv3LIwkuZIemlEnNw0bktW6zo5G3e9yclZRFxG9hQeJmlr8jvzKUlvazg09beSdo6ubJbKueZNMiqOIptwdd3oBxsOnZxQCw1RbY/cAiAiziu/0weQPa2dRC5bNWys/xbw2TKa436A8nv16bKtrlui/TmibS6n0Op6ut2US5G8hPzcrEquK1zXYtFs6a5+RjY0dtYmmlGPCbTKVO47kAe4gXutNB+yj3Y93prR0hy00ov0nIj4YRvxKnEbTaCt7D/PXLOIeHGDONVlBR5JlEE2iCwSEbUbRtoeOiPpAfKH9c2l8omkG6NBtr6y7weB3cneuK+TB7KLJ6rEDxBvTom3KvDDiLhKOY/2cHLx2NoTniu9l3PJ4ULLAK8EfhD1U6SjXPT+WWQK8nPK/533+2cRsXGDmDcDD9O7AhJN3h9JNzFv7/pQMbviDzPsuBqntcnxXa3sp1Ra2Rt9HkvML5PzuHr1cq0fES+rEesYcu3Mt1aGoi5EZgBcPCJe37ScM0H5zGxAOSGNiN9VextqxOlkURawLmMZlQWsExFLTrRvn5hLk41Zq5IV958y1rh1WUQ0GULaGo0g+2gl9orkcfclZKXpnZFzp+vGWZ88eT6f7I2CrBQ9DXhe1Mh4XOK1nk1YY9lHgXEZSBsnrWu7IaoycmsuOfdxWfLcaahezR6PswLZm9ZkPcU5wAfJRp1byNdvdbLx5x1Rc56mRpDQZNhzwz5xq+vpfgL4YjRItlZG+r2AbFBfnzx/2iMiVmtYrtafr6QXA99vMlpi0tizuFLY+oe9TZL6tghG/Yx4d5AZCefJHNb2h1bSo8nMWZs02LfXXLN1hhxaWI2/FNkq9xqyR7jWHC6NJt18tTVqZbK3cL8mvcEl3t3ksgxHUw4cQ1YyTyR/WC4kF66/hTyhOCwivtskZlf8aor0HSKi9uT2UVTgZooJhh1/J4ZIp93mCX7pCX4KOdfqaxFx/jCfxxJzafJEZ1OyV2VcL1fUyAKpzAC4UfcJk3J9yyubNPTMRJWGmr3ILMC1hpkr1/damVy+pmpNMh1+7TVny2fnz2Sj2XbAcsAiwMGlR22BI+nl5Pd5MbJ359SIuGvImIuS72unoehq8rtY+6RSI8wm3JZRNER1xV+JfI/mAo1GbpU4W5GNT38ik5h9BViBbNzcp2ljvXL+4OPK1Rsi5503ibNhVJIxKbOiPgO4NSJ+PfGefWO2upyC5l1P96t1K79d8f5Jnuu8gzyPjSHPn0ZRsT6NPDf+Idnw/+MYbu3osdizuFLYvdbOONEsYUhryon9beQb/iu6Tnaj5rp9pYfiVnIYxCsjh6Z2tjX60Kp3hq/lyGUvPh0Rn68ZrzrX7LsxNtds6AO5chmBQxjLHPmJiLi3QZzW0813xV+NsWEuS5AV11rr2HT1sm5LZtfdnvzxarJO0VXkSfPDyqG895Ap6xvNRVXvReE7n+//AL8h16Zsmuij+/FWjQbp5sv7WxXAPRHRfdJbN+4i5NybzlIK15AnaLXmKmreYcenksOO2/i+rNlve92RCqNqZW+jl0vSZRM1YPXb1oSklSNimPmKbZRhnags3aHew8zPiIg/14z7fbLh8Yqu2zcD3h0RtacASLoyxuaazSGPPWtERKMkHH0eZxtyLdcD24zbsCwPk4mjOktFjDtW1m0Qbtv8HhnVxCgaovo81pp1j4eVfS8me7eWAT4H7BQRFyjzP3y97nlZaVSfUET8oma875ONv1eVxsdLyKGL6wKfi4ij68QrMVurJKnF9XQrMd9InoMtSZ4rfgP4yRCVwlaXRavEXZo8Zu9Jri97OvmZqfUezxN3FlcK7yVfxIl6FqZ0rZ3yA9hZNmMjcuHwr0fE1Q3jXUIm1zmCHNP9iijLUTTtKezx4xDAveS6h7V7KZTZX3cjfxC/Rr4/Vw7Zo7ACOZl7D3JeyqeqFeIG8ebbQtxl2M/bY4gU4KUS9zzyc/R0MmFBrRTkww4frPlYc8jW7JNjiOGPXTFvjYg1Guz3sx43L0/2VMxt0lOhnNtyBplx9tfwyCLk25Dz1wb+fqvlYccTPMaywHrl6m+G+e5UYnbmR+/JEK3sXTEb93JJuo78fvSaO/nVYXsKKxXivYAnRsSqw8Tril07+YGkz0TEAWp/mPlVE31nq5W7mjFHduyRtAn5nuwO3ET2rtdewL5tkvpmLq/bIDxbjaohqk3VRidJ11aPNU0qT8pM4d2CrDSsFhG15tlJujoinlT+Pxx4QkTsU0Zc/W9E1E7i1nKl8GZaWk+3R+x1KFMxyN/Ad5ON9HWHW99BdnT0nCjbRiNK6cF9MTkKbvmhflMjYlZegEvm0+Os3EKMRck5a3cDrx/2+ZLrAF5HDlt41Px6LQYsp8jerc8DvyfTMu8OPLphvH+QCzS/mxzmMu7SIN7tXTFuHyZeJe7Typd6pXJ9I7JifFuLr+3SwL4N9rufTMBxBVlhv7/y/xUj+hy8psVYrb2GJd5mZMNHk33PJofIdt++PTn3sU6sFYDXkdkYryeHH7XyXMmK74nAX8iEXJeRw/hOIOfitvVarlnz/ut0Xd+SzDR7K/B3MmvfcjVj/qzfpeHzWpys+J5Ojvj4C2XOawuvWfV9/x1wVM39v1n+3kMu//NiMhkCwI1DlOuGJtsmifkQ8Ndy+Rs5N6zz/18bxFufTLZ2bXnuryeTabTyee56rM1bjrc6ma259bLWLMeplf8/2rXtx1NdvgnKvFJ5r89v+/dgyHJd0uv/Xtcbxn86ubTJBWRjWd39L6v8fzbZADXPtpoxD5/q171BmZ9MWY90mPd4RGVbDng1mUvhNuDoYeLN2uyjTFBrbyVwV8swOUm+SZxFybl1c4G1yLW5mg5rfeT5RsSFZUjPMeTQ1P9qGBPlRPuDgceXm64FjomIk5rEi/yUnwOc0zXX7DPkyVBdRzLWgtSdFr5JN3nr6eYlHUn25l1GZsP9Ptni8yGyV7duvO5hvUE5AYxmiQ/m+5yqiPhsm+FajEVEXKycN9vEqhHxkx4xfyqpVi9FZKKj44DjKsOO75J0LQ2GHXd5B5mufvUoQ/VK6/Cx5FqA7xw0kCaZH00ONx/UoUCvXq73kb1ctT/fEfHsuvv0I+lkct7Nj8kse+eQlaKfDxGzV/KDdaJZ8oMfl78rMzbM/OjSM764pEdFszk5F0l6dXRNG5D0SsYSnNQSNXs2BnAduRbeLlHmOJbhYq1Qe4uad+K1kf2wE2sxsscsyJPbpkkq1qv8vwPwtsr1oRY7l7Q2Y8Pqr43KMOeacRYDloqx5ULuIjO4nkqO9pguRrFcAZK2I4/RAXyo12/OgG6T9Hqy8XtTcg5bZ85i7eVMirslrRc5z1NkQ+OLyMb7/aJGXga1uJ5uP5Ej365UsyVnWq9rlN+D3cjjwqbk6KMPkI2YQ53vzOZKYZM1YCZUviTPJ3+0NyUrCruRrblN4n2ZHEZ3FvDeqEz2bejS6pXIDI+vUGYx+kDDMu5DztN7EznWvDMU7khJ1K0YKrOtrRgR15Qy/gf4nnI+5GFNyhhj68z1erza6xVF6eqXtEI5KW/Dc4GnRCaEWQ74AzmHr+l6a70qqWsBR0h6T0ScUidYlPkSbf1gj0KpUPU6GIocNtTmY608wWMNYiFJi0bX/MFyEtP4eBy5NuVRwFFl2HHt9S27vBDYIioJniLn+B5AtjoPXCkke8EnnB9dU+ek8zXkCf5xjCVTmi5zITYke1WvBa6LXDh62LLdxbzJD17QJFCn0haZmOAs4KzKMPMlyQWmaw8zJ38LTpP0UsZnuVyErNBOBy8iK20/k/RDMhHJUCdtaj/LZZsNACgTJnUaGG8hk5isplzj7Yiony6/32e50ee8zI/6Avl5uYx8TzaW9GsyB0KtZWHIBu8fMm8j+vZk79nrGpRxJ+Dt5Bzmzlzwj0bXUh91tN3oIem55BSh+8j39n8n2WUyryQb3LYnM3D+pdy+FTBP0sIBHUyOQoGx6VFrk3NAPwn8vxqx3sSI1tOdwEvI6Uh1bNdyGSCHu/+I/P37YYPv8IRm85zCXskugPoZkLpahk9hrGV4mJTrDzOWorlazsYpmid4nMXIVtNvNtj3AnI4wc1dt69FZvzaqma8U4DjomvOhKTnkMMe656k9HqMcS25EVGrJVe5FMOXyIQoDwO7R8T5Q5bp1xHx1Mr1y6LF5BaVuMsDP42ac3Im+sEmT/ya/GC3TiNIDT9BRXN5YGsy+2Gv+RuTxXwH+YN6UOd7U74vx5A9Xe+rEeuF/bbHEMmyJF0RE8wXqTs/TC3Oj+70RKnlZEptUyaK2IscQnoX8ATgydE8OVOryQ8meaylgBc2HFWApGdTyXIZEee0VriWKNdu242xz8+Xyd71H/fbr0ecUSy30nb2w0+QDYVvrPT6L002Iv0zIg6uGa8zB3ch8oR8L3hkndVGc3CVGa5vBt4XY8vCiGx8elxE7FMz3jURscEE2x6ZJ1cj3qvJhqi3MrZG3GbkFJwvRL11rUemnDfeDlxOj/PbmOIkRTDPPMqvAb+KiE+W63WXO7o0yvzE7rmKbc5drMS8LVqYA99COZaIljLyzxN7tlYK26RcI03AScA3IuK2YQ7io6b21gDsd+CdcFufeBMerNUnicEAcVtryZV0BVkRvE6Zte9jEdE3McAAMf/C+B7lZ1Svt3kgb3KgbPsHe34astGju6L5MJlI6aLOsKSGZTqIPLlYgjxu/J2cF1Zr+Ghp6Z9IxBDJsjS27mOvXpSfRYN1H0vcRcnv4ZHk52noxB6VXq69yIQ9TXq5RkY5VH8u2cp8e0RsPUSsVpIflFibk/Or7izX9yF70m4B3hMNMvdNd5JOjIj9um5bnnxv9oiIbWvGG8VyK21nP/wtuXZndN0+h+zJXq/3nhPG+zl9egSjwXBsSb+dqBz9tvWJNy5py6Db+sS7Bnh693dCmeDjvCYV4VFQy0mKlIlr+r3Xtc9NlEkPn0uOprgF2LbTQFj3vVGL6+lW9ptoeLGAy5v22M8UrhS2pO2W4VFQy2sAdvdwDbqtT7zfRMT6E2y7PiIe32vbJDFbbclt68DTFXO+ZJuTtC25gG3dE59Wf7BHrcVGj93IOYDHlusXksMXg1zo/FtDlnMpyGGZw8QZBbW87qPmnR99BnBCNFgqZJLHGaqXqxJnXUrlq2ljVI+YC5Gfm4+0FO/JlOyZEbFug/0vAbaPiD+V34ZTyGQcm5BZUmt/Z6a7No7XPWKOJMtlWw0Ak/yuTrhtfpJ0Q0Q8boJtTSqF55JJeS7sun1z4OMR0Xfphh7xWq1kzm9qkKW47Nf6uYlytNVngTnA9yLi1ZXHemtEPLdGrNbW063EvIn8jZ8oW2hba11Om2VwqmbznMJWRcR1ZFazd5UDz1zgQklDtQy3RePXAHxLjK0BOEwX9BNLz9k8Dwc0adX8raSdo2uMvnIsf9P5a3eT67itTJ7Q/5bhEo+spPGJXMZdjwZrzkx0YO0cyIG6rXtX0nvY4x/IdRrrGllSpjZN0Oix9hCf8beQr3/HIuSQoSXJIcS1K4WSdiEztt5SvoPvktTpnTk4Im6qGW8OmW3znnJ9ETJT8RuHOVGJiLWa7ttN7c+PnrSXq2HMVchGvb3IYa4fZvi5mY+IXOfzAHLIWRvxriTnOL29YYg5lZ6PPch1x74NfFvSZS0UcTpaQtJTmPiEb+AkF5V97iOTZZygXNR8TzJxz1DLrUTO2f4g8MHSADCX/A7VbQC4RtI+0TXHX9Le5LzcWiR9KEoSK0k7RPMkJlX/K+ldwPurPZqS3knOYa7rLcCpZZRLdX7rPow/pg/qr5I2jojLqzdK2pjMhDvtqIUkRf0qfaVSU1tEfL+M3loqxq+FejF5HKqj9cp4W5W+XtRjGZwRPEbTRGG5v3sKR6cMsXtGWz09Q5ZlFGsArtlve9Rf4Hp94Ptk2ujqgfxpwPOaDJEqcVtrydWIF+7tdSCPiENrxuh+XwK4NyL+0ev+A8T7Mpn6vtcP9voR8bImcdvU1ejx3UqjxzBzey6KiM0r1z8dEQeV/y+ImnNmy35XAFtFxP2lxfR/yPf6KcBLIuI5NWLtSba4/oNs7HgP8BXgIvK9qn2CW4ndWlY3jWB+dJu9XMr5QnPJxqNTy+X0UZwcqOGclErr9SM3Va5Hw57Cq4BNIuJB5Tyx/aMsfKwhhutPZ8pcAhcxcQ94rVEUXbFXLEHuLtfXrPsbWPZ7TkT8aIJtu0fEqTXjrUqefP6T/F0NYHNy2ZQX1O2t7zdkrynlHMcvkolCLitl3JRMYPeqGEtwUifmSsCBVOa3Ap+OzERaN9bTgZPJxsDqa7gvsHdEnFc35iiod5KiPaJ5kqI5ZAVmVTKhyVXld+twYPFoMGdPI5wLX+KvQJ7vNE161GpG03Je28ljcS85HPzQiOh7/jxJzPMi4unl/69Uz8GG/U66UtgSjS3N8ARKdkaGWJqhz+M07nIuldRnkx/Oncl1614JnBmZjXTKlWFmezH+QP61aJ4+uzv+SmRr1FxaWjh7WG0fyHvEX4LMmHZLNJgLN8kP9iujhQXNhzWiRo9+Q5p+1/BE/PIo8/EknQBcHxEfLdfrTrK/CtgtIm6QtCm5kP2eEdE4bX0ldutzNdrU9ToeC9wdJdOwaiZqkvQA+dq9OSIuLreNZE64pFsjYo0G+3UvG7QQebJ2KLkO1osaxDyC/B24B1gD2DQiQtLjgC9HRKOegOlMLSefKL+p7wYOIt8TkfPWPxU1kkZ1xXyInFO+d3eFbZjvnnL6wJNKGa+OiLMbxmm9UliJvS75W9Up4+8kLRwtZldsSpl1+kAqryFwbEyvKUJtJyk6kVwf80JyTdhbyEb6wyLiuw1jtjYXXtJW5MiLP5Hr9H6FXLpsIWCfiPhhg/K1+ttXGkV/SZ4rdZbBGXbe8SPHsR5lHOoY5+GjLVDLSzP0iL8JLXQ5l5aT6hqAO5EtGE3XAGxdZKr+pqmOB4l/F5m6+FOT9XTOR62lmweQ9Hwym+WfSsxjgT8Ca0l6W9ScbxWZXfQlXT/Yb4uI3zUtY9si4mBJhzDW6HEksLSk3Wne6PEr9V537TXk+9WElGsc3k+mqv5MZVvddake6PzIRMQlpWd06Aphp5wT/N/r+lSYUxkmsx2wf2Vb3d+1x5K98/9TTvxOpfkaXGji5Ayi4ZqwEXFvib0Q8DJyeNxlwHOjLOHTIOYHJZ0NrEIuOt4p80Jkr6tN7hBymPrmUYZ+K+cCHifpjRHxiQYxryAbti6Q9KYYnyRrmO/e3UCnAvPHIeJ0pkyIeadTNJpCUdn3d8DvOg3Ykt4O7EJO/5gyGktQ9K6pLMcADifP6Y4DvibpG0PG24xcHuthZUKve8jkco0rwhHx8iHLVPVp8jkvQ57b7hQRFyhzfHydsq5iTW3/9rW+DA4jWBamw5XCdhxADsO4uXLbOcq5QqeQWUlrmaDLWdHSQsul5e0M4AzlGouzTpOhPSPS9oH8/WSilWXIVP0bRcSNpZf0bDL9ei3Kta42IHvCAR6WdEtMg/T/HT0aPXYkvz9NGz3eCHxX0l5kYw/AU4FFyV7JJo4mT+b/Sq712OmZegpwR81Y3Sdkj9aQ81srYoL/e12fCl8HzpV0Dzks7pcApZerVs915HzM48gT+dXI7+Jdkq4lh28fXrNsRzXcNqHyeX4F+Zk8D9i1jUaZiJhnvlY0HKY/Q7y15Xj7ADtEZc3acqzdm1yiqkmlMCKXXTkXOFnSzsCBkXOja3/3lNMnTid7fK4gT0ifLOlW8nNUd0mhzzO2Fm71/9okrROV9W6VGb33IkfOLE/2zL2lafwW9VyeZ7opjRCf0FiSou8Cj5X0NpplKX4gSsbxyLVgfzNsz6gq8+rL9XcxNh/84Kg3r/5RUZaRkfS+zvEsMjt80yK2+ttXGmpP09gyOG8EVpZ0HA2WwSmWLR0HC5X/O0NyRZ73Nebhoy1Qy0szlP1a7XKW9DMm/kBHRAy9wGY5cdkQ+H00GLc/22nebHPvIufH1c02Vx1aMG5NuSZDCyQ9lqxc3gFcSh54ngI8Bnh2RPyhTrxRKMNIPksmYbgSeEVEXFu2LR4R/xwidmfYFbSw7ppyjs9KZHrrzhIfqwALR8StNeL0m98aTYevlditZ3VrW3nPO71c/yi3rQ88OoaYT1mJvz4wN4acJ9wG5ZzZB8lGhXk+IzHkPJzZou3fQfWZe9lv2yQxq8PXHgV8gKwk7UOu41t3+NoxwANkZsfO8WYhctjd4hExZb3Ckj4TEQdI+iA5EupWssHnNHLd1tbm9UpaDvhLpUe8zr6dtRlbS1A0v2iILMWV3wEY/1vQmQ9eu7KsdufVj2JJilZ/+9TyMjhl/76j6YbpjXWlsAVqeWmGst8LyMrB1mQX+CnkIqlNl1LoVYatyJbTu6KSUKNGzOPJuRNXl9bI/wMeIlv4Do2Irzcp63QnaUPydduAPMG4hkxz3SsT6yDxHgesHOMnNG9EngA+MyLm1IzXWWNuIbLn7FmM/Zj9LGquMaecV3BZRBzddfsbgKdGRN+F4+cHSReTWRh/ATyfTE4w8I/L/CKp73yyOpXCSR5n84i4aIj9W00iNZ2p5cQH6p39txqvyYnUiX1iRgyxJuVs0vbvYL8TzyFOSudpuJP0LDLD6YoRUatnTrnG3kbdozpKhfPKmMLlFCR9MyJeUnr8ryN/875feqWGaQR/F3Bq6TFalDyH2phsWNkrIn5aM97IEhRNZ6P4HVC78+ofIpOYiUyc1Mk0LmCxiKg9DaDt59z0ODBVXClsQVdryrhNDNmqXulyngtsSw79a9rl3In5THLh8UWBD0XEWQ3jPLLYvHI+17MiYjdJjwHOatAjNVGlqnGrVIm7E1lhqFbiPhpdS18MGGtXcgjYh8kUyiKHFL6drAif3iDm94HDuyuVyrT7746I59WMdzMTrzFXO+WypOsi4gkTbGu0fmTb2molHLVKhaH63gS5XMpKdRsAumJvwNiQ8/siYrNhyjrBY8whk9mc3HbsqTJJq2vtCtf8rlBLWjkihpkjNiu18TtYOSmdZxPNT0p3ix5JPEpP12ui5jqX6pN4qd+2+UFlzrbGry27LTkyZXsyGVztKQqSrgY2jIiQtH+Juz2ZyO3LEbFFzXi1R9hMhVJ5nWg+c0TNTM+9GqzL7f8P+EM0GMJezvO2JitwNwEvirFpFI1G1o2ahshoOqpeZkmPJ+fSd87NriWXFRpq+L/nFLaj9Za2TpdzGRZ1Mjm3oNPlfBg5X6FuzOeQP4L/Aj4YET8bspgPVP7fAfgmQETcqWbjuR8mD2hfA75HzhUaijLd/GvIluCLy82bAR+RtFpEfK5myPeRc0hurtx2uaRzyHkbtSuFwFq9ehkj4qLJTjB7iRbXmCv6vQ/DrHPZpuq4+nmu1+3tGZWoDOUFkLQW8DbyhOVDdeOVz8fccnkQWBPYrOvzWZsy4+yBZCryM4CfkBkWDyXnRC4wlcJ+Q22UiWfqeguZMfn85qXqT2PL7OxF/v6sOqrHWtC0+Ts4TCNOH7cqswk/8jDAPRFxG83WuFxMvddmFFkhnjJRknhFxEPkGoxnKROaPI9cD/b3ks6OiL1qhn6gcgL/HOCU8hjXlh7SBVK1F7mliuzRZM6Dbv8s23ZpGPMy2plX3zr1yWiqXO+zbvKaVYGPM0EvM9kIUreMTyMTTn6uXDpTen4u6YXRY674wLHdUzg9td3TIekisjfiSHKY5zhNWiuU8zM+DvyebNl7QqkQPgq4aqLepUliPoE8wd2F7NH7GjlnqFFCkzJ05ukxtkhz5/b/IjN91qrQ92vJatrKpf5LH0y4reZjrEvpRYqa81wk3UhWBubZBHwsGizN0La2e3tGTdJ6wBFkmu+Pk63XtdKuSzqfnFR+CnnS81sNuTZjJfbpwJ/JY8V2wHLAImQigMuGjT8sSU+IiOvK/4tGZi3ubNtqqB/FrgpXRNSqcEk6mPyurUImCPt6G6+ZMiHY80u5NiUTfOwG/CLKXDHrbxS/g20rv6vdlie/f3tG1wLqDeM9Ihokr1PLUyj6PM5SwAujZsZsSRcAryKzrF5PTnPoZIedcORLn3g7RMRP6uwz1do4h1T/ObPj8hXUjNvKvPpRUE5F6WQ0/RxdGU0bjIBrvZdZ0lnkaLefd93+THK5kJ0ax3alcHpqu8tZ0s+pLHbcFTei2WTX9cmlD1YBPhERJ5bbnwP8d0S8uW7Mrvh7kMspfDQijmwY49qJKn79tvWJdzmwS/eBq/TYfK/JEFdJXwfOiXmXPngl+TruUTdm2X8Vck3GvcjsaR8GvhMRV9aMM7JJzbNNOZk6gkxc8zHyR+ahhrFOJ1sHz6D0TKml9fWqP/hlaNc9wBoR8bdhY7dB7a8l1XqFqxwT9iyXxcgEGqc0Gd4j6WTgGeQIkVPIucI3tNEAMJuM4ndwfpG0GfA/EfGMKS7HKKZQbA7cFiWzpXKZr05Gyvd0N+oOEG9LcqrNisDREfH+cvvOwMsiYm7NeCNP1Ne2liqFrTdYq+XF4dtWHVLdfY7YpII3okrhbyJi/Qm2DTWlx5XCaUqzd2LzquRJ1AvInopTyTmUTdaZQ9KvgP27W1clbQx8PurPLdiNPJn/EPBr8odic3JI79uiwYKuZYjaaeRw3F+XmzcjW4ZfEDVTQJchs3OB1cjX71TgdJ9ATj3lHKTbgB+QSZnGiYg31IzX6dWaCzwOWBZ4TkQ0XUexE3daz9HU+Ay743506/4Iz48KVxkadQKZ8KP2kMPSGCVyeaNvRMRtbTUAWNI0WSS9n+nwPSyfxV2ja4h6GQp/etRMZFb2vQTYPiL+JOkZ5Pfw9cAmZG/9i4ctd+Wxas/B1QgS9Y2Cxk+hOIquET5RP2lW6w3WbTfolf3WJhtagxySeuMku8y38o2il1n9k1sOdYxYYMdWLwBumAkVP2USl8MY+0IOk8TlXLKF/lRgP3JMN8Aikpav21pYvJlci/FLjK/E7QvsXTdYRHxX0k0l7uvJE7WryXTPtYb1VGL+Edha0rPJJT0AfhDNlz44lhwatVeMjddv3PpTWm0nEhHxlaaxZ6FWh7FGxH1kZeOE0riwB3C0pNUjYvUhQm8sqbN+mYDFy/VGCQtGoM21pDYkG6CuBa6LiIeG+b50aGytzD3JIbjnAo2Wt4iIjcvwpb2An0q6G1hK0mPqNhrZGCkXSSdf1ylfJL2f8v2eDq34C3dXCAEi4ubymW9iTuX3fQ8yYca3gW9LuqxhzEd0Dwmn5hzciOg01naG6HUSFL02GibqG5HqHL9zu64HOQ+tjkPINfZeSo8G64ZlbG1xeOXc9y+UMl1W9t9Y0q/J5dzqrsEJY7991d+9TtkWaxDvcElvn2Bb017m1ZVLzXQTQ84vd09hCzSCrJmj6HJum/okcSGXz6iVxEWZNbM6tOeRTeTr2DQ99cpk0ownMVaJO3ZBPZlSLlLf6T1amVLJblpJkPSpXjeTPzirRsS0blyarj0Akh5Nfq57ZS8cNvaasQAtG9FN0l1kb4LIk8hTOpvIBppaJ/eVCtcewF1kRrcnNzlGSNqB/O49F7iwlO27bb7PZSjhXsCLgdsjYuu2Ys8G6r1I+hkR8ecpLRiPHG+7T8yWJzM2HhwR35v/pRozoikUVwGbRMSDyqkz+0fELzrbotl6j60OCVf7ifpmjK4G66HW6m2zJ065VM/NwPtibG6iyPfpcRHRr0F7vhhFL7OkvsuARc05uONiu1I4vNKSNWHWzCYnZ6Pocm6bWk7iMgrqsXBoCzH3Bd7A+FTAx0TESW0+TlNdB93VGFumYAlyKG6vbGKDxhbwUjJj5jXkj2OryQXa0N0DULeSMEqSXkfOv+ksVfN3snf9MzXjnNFve0Q8v1kJp7+R/igOWeFSzj/6GvDthqMbJov/SGNC+Zw/IyLObftxFkSaD4ukD6vHZzuAe4GLIuKulh5jXfI3Yc+6Fa4RTaE4AtiZMncZ2DQiQrkkwpcjYpua8VodEq4ZkKAIZsaoHrW4OLyk30bEenW3TRW1tBzcKLlS2BK1nzWz1YnNGsHkXrWfxKVvC1GTA2/bczDKQfeNwJuAS8gD2abkj8Unp0PFcKJeZmVioLkRUXsImzKj7H7ksNlfAR+OiOuHLWvbpnMPAICkd5At/gd15j1IWgf4JPCriPhAjVh3k/MTv06+J+OG3izIFQVJHxqmcWPAx1iE7HX86igfZ1CSDiBPvB9dbvobDRoTZrPynbmelhZJn9+6f7tr7ttK4rESa2Pyt6Az+uYqMvtooykUJeZWZNK6H3d61ctv1qPr/var5Tm4miEJimbCqB61uI6r+ifCmTaVwrZ7mct0qH71g1c2ju1KYfvUTtbMVruc2+yyr+zXdhKXfl+URgdetZ/F9QKydfXmrtvXIjMLblW3jG2TdDvwPxNtj4gJt00Q70DgYOBs4CPTcVjiTOgBAJB0PbBxRPyr6/bFyfTcPTOKTRBrDrk+6FzyJO8HZDbTq1ss8rTUZmOPeq/JeCCZpOHyiNi1jccZRpuNCbOZWl4kfRRKGXcnP48/jIirJD2PTJO/eK8Gv0nitZp4bH40yLShzSHhkzzOdJ2eMCNG9XSUz/2eETHwGriSvgz8Dnh/VCozkt4JrB8RL2u/pPWMopdZ0ot63LwGOQd0TkSsVjfmI7FdKWyHWs6a2RV76C5ntZitr7Lf08lFrHsmcYmI8+rGbJtazuKqEaxT2DZJdwDHMfFzfl/NeA+TP6p3M+9cz4ejQba5ts2UHgD1SRetButnVfZdlDzxO5KcX9GrxXiBUXoCnsXEjT0DD9vUNF+TEdptTLCksUXS9wK2AZoskt465Typ1cn5qFuSyzI8jVx/7LsN4j1AfrbfHGOJx4bpNZvyDKh1KZe8mEtLc3Cn+fSE1kf1qN3snr0a4Q4iG+Euq9MIV2J9kRytdVkp36bkKK5XRcRfmpazLaPuZS6Ng4eTw6U/AXwxIh5oGm/Ku5IXBBpN1sy2u5zbzNaXO0WcJ2kL8gu+HzySxGWrpq1xyvmIezF+vt7Xmr6GtJ/F9Z8Nt81Pd9St+E2iV4uyyJbn6dJi/BjGegCOLr3Oi0t61HTpAShul7RdRJxdvVHSdsAddYOVyuBzyee9FrluaN0MczPRE8iGqJ4NH0CdE951YmxNxi8wzdZk7OiuEJbb/lkabaym8np+C/iWyiLpU1ykjs3IpUseLhXXe8ikGU17uB4LvAT4H2XStVOBpllCAeZIWo4WGmTml4i4CLhI0pvJk+dGJpie8JZWCtmCrlE9Ow47qkejye75FcYa4V5Fvn6LkMucXFYnUHn8lyjnyG5Qyve2iPidmmfCbVVEPGuibcOUUdITyTWPn0I2Br+2jXMd9xS2QCPImtl2l7NanNxbidlqEpfyIT8H+BFwaSnbU8ghcttGxHUNYjbqBe0Tr/M6zrOJhq9j29p+zl2xNyF/FHcHbiKTaUzpYrPdKj0Ac4GnM016AAAkPQk4HTiP8b3r25A/igMP/SxDZzYEziKHLl/VfomnpzY/420Npx8lSWeTo0S6GxO2Bd4ZEc+empLNLGp5kfRRGOXnUS0kHpP0b+D3TDwSZVqMzlAm7DmYbEAKhkgIN4OmJ/Qb1RNRMzOsRpDdU9KVlUa4ObTYCDede3A72iijpG+SFfWjyEaecWseD3Mcc6Vwmmq7y1ktTu6txGw7icu3gFMj4tSu219ErrnXaxz1ZDFbzeI6itexbcP0Tk8Qb33GTiTuBb4BHBoRfV+L+UnSD8ihzKdHJfV/ael8QQyRjbJNymx6jwHWZ/wSKb8Ffh8Rv6sR62Gg81x7nQBM9VqCI9NypfAh8nXsHGMXB+5nGr2OkzQm7DabGgSGofm4SHpTXQ2P3Y241D2x7/M4jRKPjbLRsS1qOSHcDJqe0Or5iUaQ3bONRg9J61SHsE7QgzttEsxBu2WcoCOq8/s1VMOMK4Ut0AiyZk7yeI0nNktaFuh8kX8Tufh103K0ncSl33yrCbdNErPVLK6zUal8/JIcLnJDuW1a/SBK2pWsuG5P9jZ/HThzmLH1oyDp+8Dh0TXhX7kMwrsjYpfee1qVpP3IYUjLRcQ95bZFyGHsb4xpsBxO20oP+F6MDZO6Gjg5Iv49pQWbQSRdHmUOtKRjgbsj4j3l+mURsckUFo9SjvXI9WVv69q0JvCHzjG4Rry+w2IjotZw8xlSKWw1IZxmQIKiUdAIsntWGuEgj2O1G+EkfSYiDpgJPbgzoYxVnlPYjo/32RbkQWQo3V3O5I9Gnf0XAT5HLt56E/kFXFPSaeRY5CYnz6uSz32ieT11n3e/xZ2bLvx8aI/bHsni2jDmbPMissL1M0k/ZGzR8GkjIk4HTtfYgsX7AsdLOpPMyDld1vxcq7tCCBARF5cTFhvMv8i52/+Q9FvgPWQl8SIy497ASmXrtcDjgCuAE6bbiZ4yYVb3iBHI+bP/JjPwHdE9vNTmMUdj84y3A/avbJsu50OfIBuOxvXqSFqxbKvbcNTv/kH9OcifrHn/qbB0d4UQICJuLqNHaomIh8hh+mdVpicsCfxe0nSantA5TsyziWajHv5X0rvond3zgiZljIg5TfbrsmL5+xrgOjKxXqcHd7r1dO1P9jKPtIxlXuWeZO9/rbVHq6bLQXBGG+V8jgm6nJtMbH4HObl89c7YbeXk+mPJ8eHvbBCz7SQuK0l6U4/bxdhBoJaI+PUjQcZncX1tTMOFQ6ejiDgNOE3SkmSjwhuBlSUdR85J+fFUlq8qIv5JDm/9hqSNgC+TFcQ2fojasFifbYvPt1LMfO8AnhoRN5SRGv9H9gyc1iDWl4H/kL3hO5PDeg9uraQtiIilJtpWejE2JIdPNz4ZmCW+Dpwr6R4yMdgv4ZFh3Y1HzbSs1YajiHj5RNuUiWfqemYZejvBwzVfI61FI0sIF9M3QVHf40RDryeze94g6TK6snu2/Fh1dM45Vmb6J5gbWRI89V57dO5QMT18tB1qOWtm213Okq4CtoiI+/9/e/ceJFldnnH8+4AouCKgBgsRJKwSVG6KRAQlMeZmvCHInShqFA0mEMUoYmlKo6WikiiCgJEgiBBU0Fy0MCAgolKAy24AL4kIqCkpCXIXEZ788Tuze+jtntnuPj2ne87zqaLqXKZ/590ZZrp/t/ftuf4o4NujjCw0vYxE0rvnuz/s3odau40WDh3wjK0oH0pHqks5ayQ9hpLR7oCGBwbGUn3I2Z8yYrYFcC5lpnBFm3HNkfQ54CLbp/Zcfy3wx7YPaCey2dJnX8o45TzqiQ8eBlwx7B6XaSDpcNsntx3HtFODRdInYYElewPvDdH+JpTVHwdT9lFuOeTrJ1YjrSlqOCGcZiBB0STpodk9r3WV3XPUbUyToCktMVPXVIxquPboQ9pOp3B8mkzWzEY3Nkta6QEb1OsfioZss9EkLgs8a5lrCUSGeF3jhUNrbT+O0jE6iLKU9jzb/ZarxoTV/khuT1kO9Tnb32w3qrVVndbzgF9TEoZAySL2cEpCnEYLKy9Vkn4CfLR26c31c9sfXetFg9ua+uyj0R2TGDiqLas/mDLTszFl1celrrJKjhhrozXSmqLmE65MfYKixdC7jclTmN0TVq+C28dTkmCun3FiVMO1Rx/SdjqF49NksmY2urFZ8xd7/rpHKECuCSRxkbQlZRR3pe1fS9qcMgJ5mO0njNDexbUYm8jiujFlKe/BlAyS51Fmy1ofHe0ySadRZtT/c5wPOYtF0vNZs9TvWtsXtRnPrGlyVUETiQ8imtL0wJGkz1I6bRdQOjMXUbZ+jDyroLVrpJ05Rcv1GqcZSFDUJM1Ids9ZmMGdRIw9ExJztUcPs73V2PGmUzg+TSBrZk8bY085q6SwfZABSWFGGWGQtGufy6uTuNjebcj2jqK80fw3Zd/fP1JG/z8DfMj20MW9F3je0MsfJN0LXEHZ03SZbTc1QhPjqQYQjqDsCTNwHXCi7Z+3GlhExJCaGjiqBoRFeR89x/bNY646mliNtGlVbb/ZxfZvVLKuv972pXP3Rtl+M800I9k9Z2EGd9IxqoHaow9pL53C8c233KjppUjTOi2uhyZxeb9HSOIi6TrgudUvz9aUzuFetkfKcjXgGWMtf5D0N5RfwGXAWZSkJl9Lp7Bdkvak/Dz+mTK6PleX6lXAIdO4lDTGI+mFwDGUvS5zgwAftP0frQYWU0vS9nPbOSQ9wrVyHpJ2b/K9ZppI2p7ynncAJev29sCOoyxX1wRrpE0rScdSklD9grJ/8pnVgPCTgdNt79lqgA2TdK7t/VQSMn2PKa3POAszuIsZY7U3+hjPk1xqwTbSKRxfn/0tq28BR40ypTsL0+JAo0lc+uztaWwErunlD9VeioMoHcSnAO+mjND8oIl4YzgqdaneaPu7Pdd3AU62/exWAouJqPaQHk5ZlXBldflZwAeAT9k+pa3YYnrV32O6updUpSbqwcArgJ/Y3qPlkBaNxkgIpylPUNQkSa+zfWrT25iaNgszuJOKUdJzKLksLrV9i0q29bcDzxtnGWk6hQ1ocn9Lrc1ZmBZvNImLpFso/845B9bPbf/1CDFOfPmDpB0pb7L7217eVLux7iRdZ/tpw96L2VRfVdBz/bGUZd1Lrnh9jE+1jNnqyZ7de77UqdQu3t/2mUO+7tC510jas74KQ9KbbJ/QcKhjURLCNaKJbUwTiGnqZ3AnEaOk4yg/ixWU+rr/Bvwl8H7KIPivRo43ncLJ0uhZM2dhWvximk3i8qr57o+yZFYNZ3GN6STpemCP3plfldIZl3vEcgUxnSRdP6jjN9+96LYuzhSqFGs/gtIh+jLwter8aOAa2y8bsr2p/x4qCeEmapq2Mc3CDG7TMVaDos+sPs9uBvwM2Mn2D8eNNcXrG6J5smYCQ2fNBNbXmuKWLwBeX7s39M9N0iOB+10lVpH0O5TRixttf3GE+LD9+/M8b4MR2uv7B6YaoXrJsO1VGi0cKulO+mdcTabCdh0PXCDpaEphXYBdgQ9W92JpuUPSzravqV+UtDNwZ0sxxfR7oqSPUf5ezx1TnQ9Vr2+GnAHcRlnN8xfAWymZTPf2aPVbNeC433lbbmHthHAvbzmmmbPQNqYWQ1ut3z7gadvGM4EY752bDbR9m0pCy7E7hJBOYSPUkzVTUj1rZr8Mnevic8Al1Ubfe4FvVM96MnD7CO19FXgt8MOqjW8BnwVeLGk328eMGOdqvUlcKKlyR22rvpb9Tyj//nOHbcf2A8BXgK/Ulj8sA34qaejlD7Y3rsXYqeVG08z2KZJ+BryXkn0U4Frg723/a3uRxYS8BfiySimSqygDNbtREgsd2mZgMdXeWju+sude7/lSsa2rOsSSPkW1jM32qIMnHnDc77wt76BsPzkJOEvSOS3HM6tOpuwhpNrG9AHWbGM6hbIvNRbfcklfrp1vUz+3/dJRG87y0QZoQlkzm5xyVq1AvaT3Ao+xfUS1r+Aqj1C8vtZ2Y0lcqj88BwMvooz07Ul5U7tn1PgGPGfs5Q/TslQmootU6rnNlSARZRDgEx4ho2LEUtX0Ek9J91A+4whYXh1TnW9re9k48TYpCeHGMwvbmLpIJdv/QLYvGbntdArH1+eP7lRkPaqTtNL2TtXxN4HjbJ9fna/+xR+yzUaTuKhkcb2JMrp3vu07Jd0wTlKYhZY/9CaqGLLtdAqnhKR3zXPbtt+7aMHExEna2vZNbccRs6fau34kMFc/+HrgY7Y/015UkyPpAeBu1izt3Ai4hxG3PEh60nz3bd84SpyTloRww9MMZPeMZmX5aDPqexMANq+fe4SsmROwUtKHgZ9SshVdACBp0zHafD0lictJrEniMs4owxeAvSm1lB6Q9CXGX47S6PIHSfvUTjftOWfU/Zkxtn7JnJZRlkw/lrKsNJaO8yl1KJH0Bdv7thtOzIJqUPAo4M2Uvcdz9UyPk8RS7BjaXr/h9tbq9FUZPm/1FM8y2F5FqWs69laZDml6G1M0QNLLgCfa/kR1/h1KJQCAv7X9+ZHbnuLf4ZkxiayZTZO0EWV0dAvg03MJGiTtASy3fcYIbTZew6a2L/EgSiKcTSgf7P/d9l0jtNfo8odqD9Mgtv2aYWOMZlVLg4+k/H/zL8BHbN/SblTRJM1TWiBiEJV6pgfa/nHP9W2As23v3kZck1TtpX8DZTB4JeX9f+Qac9W2lg8A/0cZbDsDeBywHvBK218dO+gxJSFcc2Yhu2fXVKv9DrR9c3W+gpKQchlwmu0XjNp2ZgobMKGsmY2yfS/lD3nv9cuBy0dss9EkLlWbBi4CLqoymP4ppYN4AmtGQobRaBZX268edE9SZitapFJ+4s3AIcDplJTNQ+9rjZkwX7KLiEEe3dshBLD9Y5XSDUvR6cD9lFmeP6PswT1yjPZOoCRy2YTyXv1C29+WtD1lZqn1TmESwjVnFrJ7dtDD5zqElcts3wrcKmmsPb3pFDasqayZTZO0irU/SP2CMrv3YY9R7BKgev3ngc/PJXEZIcZ+o3tz+yDur0Z5j7V94RDNLubyh+MpS2BjkakUc92HsiR4x1FmlWOm7CzpDsrfh42qY8hMQMzv3hHvzbKn1ZLM/RMlgds4HmZ7bvvJe+Y6Dba/Vxb6TJ0MGsVSs1n9xPabaqejTJ6slk5hQwZkzfztprNmjuHFfa49hpLC/ePA64ZtcKEkLsO2Vx/d6/Os9YEdKGU01nlzs+33SbqQNcsf5t4g1qPsLWzSVL4jdsRbgPsodamOrX04SSdhCWp6n1R0xlMlrexzXcC2ix3MIrl/7qBKGDJuew/Wjns70umARUzedyS9zvap9YuSDmfMQZ/sKWzAJLJmLqZRl1dIuhr4Q5dSHHsBZ7MmictTbTdew0bS4bZPbrrdJki6yfbWbccRERFrm9XMmeOoZR+Famad8bKP1rOZzrU11/aGtjdoIu5x9CSA+zBwdP1+EsLFLJO0OSXZ2n2UhFlQaqI/Atjb9s9HbjudwvGpFKvfG1gFnAV8CVhleyZGHscoSdG5GjZ9luGuvgVsZ/sRixxSREREVJIQLrpA0h9Q9ggDXGv7orHbTKewGU1nzWyapH719DYDDgXusj30Usou1rDp4khzRETEUiBpX9vZ+x/RRzqFE9CTNfOPbI+18bMJkr7ec8nArcDFwCm271/rRQu3eSylA/wLYGtKtkdXSVxOt73neFFHRERENCPbPCIGS6ewAQtlzQR+wPBZM2dC12rYpP5RRMTSImkrSt2v49qOJSZL0s22t2o7johplOyjDZhE1symSXoJsHJueaOkd7EmU+iRtm8Ypd2u1bCZ72cdERGzQdLjgP0oK3q2BM5rN6JYJJkJiRggncIJqwq8XyPp4y2H8j5gdwBJL6bsJTwIeAbwSUpNxYiIiCWpqqH7ckr5qO0oHcFtbT+x1cCiUQskhHv8IocTMTOyfLQjejKFfhr4vu0PVudX2+6XiCYiImJJkHQvpY7XO4HLqj3wP5qVTOGxbpIQLmI0mSnsDkl6FKWm0AuAE2v3NmwnpIiIiEXzDuBASk3hsySd03I8MQHp9EWMZr22A4hF8w/ACuBK4HrbVwJIegbwv+2FFRERMXm2j7f9bOCllKWE5wNPkPS2KkFaLAGS7pR0R5//7pR0R9vxRUyrLB/tEElbApsD19h+sLq2BbCB7ZtaDS4iImKRSdqRssdwf9vL244nIqIt6RR2hKRDbZ9ZHe9p+5u1e2+yfUJ70UVERERERFvSKeyIejKZ3sQySTQTERFLXerMRkQMlkQz3aEBx/3OIyIilpR6nVlJ37X9jDbjiYiYJkk00x0ecNzvPCIiYinL+15ERE1mCrvjqZJWUmYFl1fHVOep0RQRERER0VHpFHbH9mRkNCIiOkrSPrXTTXvOsf3FRQ4pImJqJNFMR8yzwR7gPuB/gGNtX7h4UUVERCwOSafNc9u2X7NowURETJl0CgNJ6wM7AJ+1vUPb8URERCwmSfva/kLbcUREtCWJZgLbD9i+Bvh427FERES04Pi2A4iIaFNmCiMiIqLTJN1se6u244iIaEtmCiMiIqLrMkIeEZ2W7KMRERGx5ElaRf/On4DHL3I4ERFTJctHIyIiYsmT9KT57tu+cbFiiYiYNukURkREREREdFiWj0ZERMSSN0+9XlHqFD56kUOKiJgamSmMiIiIiIjosGQfjYiIiIiI6LB0CiMiIiIiIjosncKIiIiIiIgOS6cwIiKiRtIbJL1yga85TNIJA+7dNebzL5b0rHHaWMx2IyJi9iX7aERERI3tT7b1bEl5X46IiEWXmcKIiFjSJG0j6XpJp0q6VtIFkjaStFzSVyVdJekbkravvv7vJB1dHe8maaWkb0k6TtJ/1Zp+QvX6H0r6UM8zPyLpakkXSvqt6toukr5dtXeepM2q6xdLer+kS4Ajqyb2k3SFpB9Iel71dRtKOk3SKknflfT8Ba5vJOns6nnnABtN7rscERGzLJ3CiIjogqcAn7D9dOCXwL7AKcBf2d4VOBo4sc/rTgPeYPs5wAM993YBDgB2BA6QtFV1fRlwte1nApcA766ufwZ4m+2dgFW16wCb2v492x+pzh9m+3eBo2pfdwSA7R2Bg4DTJW04z/U3AvdUz3sfsOs6fJ8iIqKD0imMiIguuMH2iur4KmAbYA/gXEkrgJOBLeovkLQpsLHty6tLZ/W0eaHt223/CrgOeFJ1/UHgnOr4TOC5kjahdPwuqa6fDuxVa+scHuqLPbECPBc4A8D294Abge3mub5X9XxsrwRW9n5TIiIiIHsKIyKiG+6rHT8APB74pe1d5nmNhmxz0HuqF4wO7h7Qdr3dQfHMF+e6PDsiIjouM4UREdFFdwA3SNoPQMXO9S+wfRtwp6Tdq0sHrmPb6wGvqI4PBi6zfTtw29z+QODPKUtLh3EpcEgV73bA1sD31/H6DsBOQz4vIiI6IjOFERHRVYcAJ0l6J7ABcDZwTc/XvBY4VdLdwMXA7evQ7t3A0yVdVX39AdX1VwGflPRI4EfAq4eM98Tq9auA3wCH2b5P0qDrJwGnSVoJrACuGPJ5ERHREbKzsiQiIqIfSY+yfVd1/HZgC9tHLvCyiIiImZKZwoiIiMFeJOkYyvvljcBh7YYTERHRvMwURkREREREdFgSzURERERERHRYOoUREREREREdlk5hREREREREh6VTGBERERER0WHpFEZERERERHRYOoUREREREREd9v9FPOGEJZGtHAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "pd.crosstab(df2.neighborhood, df2.attended, normalize='columns')\\\n",
    ".apply(lambda x: abs(x[0] - x[1]), axis=1).sort_values(ascending=False)\\\n",
    ".plot(kind='bar', figsize=(15,5), color='royalblue');\n",
    "plt.axvline(5.5, color='r', linestyle=':')\n",
    "plt.ylabel('Abs difference [Attendance Rate Yes-No]')\n",
    "plt.title('Absolute difference between Attendance Rate per Neighborhood');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can clearly see there are six neighborhoods with distinctively different proportion of attendance rate. Since the difference is absolute, we don't know yet which locations have positive or negative impact:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ref = ['ITARARÉ', 'JARDIM DA PENHA', 'SANTA MARTHA', 'JESUS DE NAZARETH', \n",
    "       'SANTOS DUMONT', 'JARDIM CAMBURI']\n",
    "df2 = df2.query('neighborhood in @ref')\n",
    "\n",
    "pd.crosstab(df2.neighborhood, df2.attended, normalize='columns')\\\n",
    ".sort_values(by=df2.attended[1], ascending=False).plot(kind='bar');\n",
    "plt.ylabel('Proportion')\n",
    "plt.title('Proportion of Attendance Rate by Top 6 Neighborhood*');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It looks like patients from Group 1 (Jardim Camburi, Jardim da Penha, Santa Martha) tend to be more likely to show up, whereas those from Group 2 (Itararé, Jesus de Nazareth, Santos Dumont) are less likely.\n",
    "\n",
    "Perhaps it would be more meaningful for the classification model to create two dummy variables capturing if a patient resides in one of these two groups. We can compare their significance afterwards:\n",
    "+ **Group 1** (Jardim Camburi, Jardim da Penha, Santa Martha): More likely to show up\n",
    "+ **Group 2** (Itararé, Jesus de Nazareth, Santos Dumont): Less likely to show up\n",
    "\n",
    ">Let's create two new features to represent whether a patient is from one of the neighborhoods in Group 1 or Group 2. This will be useful when implementing a regression model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Creating 2 new features to capture the impact of these groups\n",
    "group1 = ['JARDIM CAMBURI', 'JARDIM DA PENHA', 'SANTA MARTHA']\n",
    "group2 = [ 'ITARARÉ', 'JESUS DE NAZARETH', 'SANTOS DUMONT']\n",
    "\n",
    "# Excel like function to set value based on condition\n",
    "df['reg_neighb_group1'] = np.where(np.isin(df.neighborhood, group1), 1, 0)\n",
    "df['reg_neighb_group2'] = np.where(np.isin(df.neighborhood, group2), 1, 0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Q5) Does receiving a scholarship (Bolsa Familia) has any impact on the attendance?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAETCAYAAAA7wAFvAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAAhOUlEQVR4nO3dfbwWdZ3/8dfbI4IIigq6wgHBuxLQSMDUWGMt46ZN8bFZWmuhGbp51z7WCn9b+9vW1q0fdqsm2XqT5kqWeVNLokbomhrIeguKIKActBQSAhSU4+f3x3wPDRfXOec6eOacA/N+Ph7X41wz853vfGauOfOZ73fmmksRgZmZldcunR2AmZl1LicCM7OScyIwMys5JwIzs5JzIjAzKzknAjOzknMi2IlJ+pSkezphue+XtFjSekmTOnr5bSVpsKSQtGtnx9IVSJos6cFOjmGOpLO3c97lkj5UQEw3SPp6C9PXSzqovZfbEZwI2ijtZG+kD/2Pkq6X1KsLxLXNwSwibo6ID3dCOP8GXBkRvSLijuYKpX/21yR1rxi/1T9yGQ/UFfvZH9JBqKb9rCscyJtIGibpnvQ5r5E0X9LEzo6rCGl/X9rZcWwPJ4Lt89GI6AUcBYwGvlJZoCMPWl3wAHkgsKClApIGA38NBHBSB8S0I2raz0YA7wUu6dxwtssvgXuB/YH9gAuBP3dqRM3ogv9HHcaJ4B2IiJXAr4HhAOms9TxJi4HFadznJC2R9CdJd0nq3zR/Kn+hpKWSVkmaJmmXNG0XSV+R9IKkVyTdKGmvNK3pDPmzkl4EZgMPpGrXpLPIYyvPDCUdJ2mepLXp73G5aXMkXSrpd5LWpbO4vs2te3PrJel54CDglymO7s1U8WngEeAG4DO5em8CBuXm/1K1dUtlz5L0TDrbnCXpwIpte27qonpN0lWSlKbVSbo8bfOlwEcq1u3MVO+69Nmck5s2VlKDpH9Kn8vLks7MTd9d0rfS57ZW0oOSdk/TjpH0UDozfkLS2Oa2b15E/AGYRZYQmpYzVdLzKcaFkk5J4w8HpgPHpm21Jo3vntb5RWUt2elNcTVDkq5I6/CspA+mkadKml9R8J8k3VGlgr7AEOBHEfFmev0uIvL75MmSHpf057Q+43NVHNjc/ijpJEkL0rack9a72kocLenhVO5lSVdK2i03fav/WWW+kz7btZKelDQ8V+Xekv47xfR7SQdX1HVIen9D2sb3prL35/fPLici/GrDC1gOfCi9H0h25ntpGg6ys599gN2BE4BVZC2H7sAVwAO5ugL4bSo/CHgOODtNOwtYQnZQ7QX8ArgpTRuc5r0R2CMtq2ncrrn6JwMPpvf7AK8BZwC7Aqen4X3T9DnA88Bhqb45wDea2QatrdeWbdTCdlwCfB4YCbwF7N/c/M2s26RUx+Fpfb4CPFSxbX8F9Enb9lVgfJp2LvBs+vz2SZ/BlvrJEsPBgIAPAK8DR6VpY4HNZN1f3YCJafreafpVadsNAOqA49I2GgCsTuV3AU5Mw/1q2M/qgaeA7+Wmnwr0T3V9AtgAHFD5uefKfxe4K61vb7Iz9f9oZtmT0zr+Y1rHTwBr07zdgT8Bh+fKPwb8XZV6RHZC9Kv0ee1fMf3oVO+JaT0GAO9ubX9M4zak+boBX0r7wm5Vtt1I4BiyfWQw8AzwhYr9JP8/Ow6YT7bfiGz/atquN6R1PzrVdzMwo6KuQ3Jl1wHHp232vcrPpCu9Oj2AHe2VdrL1wBrgBeAHwO65HeGEXNlrgf+XG+5FdtAbnCs/Pjf988Bv0vvfAJ/PTXtXmrdphw7goNz0pnHNJYIzgLkV6/IwMDm9nwN8pSKWu5vZBq2t15Z/xGbmH5PK903DzwL/WLGNW0sEvwY+mxveheyAfGBu247JTb8VmJrezwbOzU37cGX9FfHeAVyU3o8F3qiI5RWyg80uadp7qtTxZVIiz42bBXymlf1sXYrtN0CfFrbp48DJlZ97GhbZgfPg3LhjgWXN1DUZeAlQbtxc4Iz0/mrg39P7YWQnFN2bqaseuJLsoP42Wevu0DTth8B3mpmv2f0R+Cpwa8VnvxIY29r+B3wBuD03XPk/ewLZCdkxwC4V894A/GdueCLwbEVd+USQTxK9gEZgYHOfYWe+3DW0fSZFRJ+IODAiPh8Rb+Smrci970+WLACIiPVkZ4EDmin/Qppnm3nT+13J+lqrzduayvqa6szH8ofc+9fJdt5W62pmvVryGeCeiFiVhv+LXPdQjQ4Evpea/GvIztREbevTn223+xaSJkh6RFm31xqyf/h8N9nqiNhcpe6+QA+yg161eE9tijfVOwY4oIV1nBQRvcmSz7vzMUj6dOpSaapreEWMef2AnsD8XPm70/jmrIx0BEvy++aPgU+mrrYzyA7Km6pVEhENEXF+RBxMtg02kLVkIWuRVdtWTVr6/PL739tkn+c2+5+kwyT9StkF9z8Dl7HtdtqyL0TEbLLEdRXwR0nXSNqzhpiqyde7nmwf7d988c7jRND+8v88L5Ht/ABI2gPYl+zspcnA3PtBaZ5t5k3TNgN/bGZZ+ffVVNbXVOfKKmVbU8t6VZX6pT8OfCD9c/6BrAviPZLek4pVrku1dVsBnJMSctNr94h4qIb4X2bb7d4UX3fgNuBysq6MPsBMsiTTmlXARrJupWrx3lQR7x4R8Y3WKo2I+8nOMC9PMR4I/Ag4n6xrrw/wdC7Gyu21iqylMiy37L0iuxDdnAHpQN9ky74ZEY8Ab5Jd7P8kcFNr65DmW0F2gG3qc19B9W3Vmsr9T2SfZ7X972qyFuehEbEn8H/Y9rPcantFxPcjYiRZa+cw4IvbESPk9jFld3ztw1/+v7sUJ4Ji/RdwpqQR6QBzGfD7iFieK/NFSXtLGghcBPw0jb8F+EdJQ9JOdBnw04oz0bxXyZrezd3HPBM4TNInJe0q6RPAULL+2yLWqzmTyJrIQ8kufo4g64f9H7ILyJAlu/x6VFu36cAlkoYBSNpL0qk1xn8rcKGkekl7A1Nz03Yj69N9FdgsaQJZ11Gr0pnpdcC3JfVXdlH62LSNfgJ8VNK4NL6HsgvP9TXG/F3gREkjyK4LRYoRZRer8xc0/wjUN10UTXH9CPiOpP3SPAMkjWthefuRbaNuabseTrYPNbmR7Mx5c+Qu/ual/fprkg5RdvNDX7JrX4+kIteS7UcfTNMHSHp3DdviVuAjab5uwD8Bm4BqJwG9ye5SWp/q/oeWKpY0WtL7Ur0byBJ7Yw0xVTNR0pj0OVxK9j/SllZ8h3EiKFBE/IasP/M2srPQg4HTKordSXZx6nHgv8n+OSA7oNxE1qe6jGyHvKCFZb0O/Dvwu9T8P6Zi+mrgb8n+aVaTXWD721z3THuvV3M+A1wfES9GxB+aXmQHlU8pu4XvP4CvpPW4uNq6RcTtwDeBGanJ/zQwocYYfkTWP/8E8L9kF+Kb1m0d2S2Ot5L1fX+S7CJrrS4mu7A7j6wr4Jtkfc0rgJPJzkhfJTsb/iI1/g9GxKtkB9+vRsRC4Ftk13j+CBwB/C5XfDbZTQx/kNT0+X6Z7ILqI2l73Ud23ak5vwcOJWtN/DvwsbQPNbmJLPm01Bp4k+z6zn1kB+OnyQ7Yk9M6zQXOBL5DdtH4frZttW4jIhYBf092k8Iq4KNkt9q+WaX4xWSf4Tqyz/2nVcrk7ZnKvUbW/bSa1BLbDv8F/F+y/WAk8KntrKdw2rob0DqSpCBrsi7p7FjM2iJ18b1CdjfV4s6Op6uRdAPQEBHbfMeoK3KLwMy2xz8A85wEdg6l/SadmW0fScvJLrhO6txIrL24a8jMrOTcNWRmVnJOBGZmJbfDXSPo27dvDB48uLPDMDPbocyfP39VRFT9NvkOlwgGDx7Mo48+2tlhmJntUCRVPmJmC3cNmZmVnBOBmVnJORGYmZXcDneNwMysPb311ls0NDSwcePGzg6lXfTo0YP6+nq6detW8zxOBGZWag0NDfTu3ZvBgwez9ZO3dzwRwerVq2loaGDIkCE1z+euITMrtY0bN7Lvvvvu8EkAQBL77rtvm1s3TgRmVno7QxJosj3r4kRgZlaDyy67bMv7NWvW8IMf/KDd6l6+fDnDhw9vvWDO5MmT+fnPf94uy3ciMDOrQZGJoLP5YrGZWYVJkyaxYsUKNm7cyEUXXcTSpUt54403GDFiBMOGDaOxsZHnn3+eESNGcOKJJzJt2jSmTZvGrbfeyqZNmzjllFP42te+xvLly5kwYQJjxozhoYceYsCAAdx5553svvvuzJ8/n7POOouePXsyZsyYLctubGxk6tSpzJkzh02bNnHeeedxzjnnEBFccMEFzJ49myFDhtCuT46OiB3qNXLkyDAzay8LFy7cZtzq1asjIuL111+PYcOGxapVq2KPPfbYMn3ZsmUxbNiwLcOzZs2Kz33uc/H2229HY2NjfOQjH4n7778/li1bFnV1dfHYY49FRMSpp54aN910U0REHHHEETFnzpyIiLj44ou31PfDH/4wLr300oiI2LhxY4wcOTKWLl0at912W3zoQx+KzZs3x8qVK2OvvfaKn/3sZzWvE/BoNHNcdYvAzKzC97//fW6//XYAVqxYweLFLf8Q2z333MM999zDe9/7XgDWr1/P4sWLGTRoEEOGDGHEiBEAjBw5kuXLl7N27VrWrFnDBz7wAQDOOOMMfv3rX2+p68knn9zS/7927VoWL17MAw88wOmnn05dXR39+/fnhBNOaLf1dSIwM8uZM2cO9913Hw8//DA9e/Zk7Nixrd6OGRFccsklnHPOOVuNX758Od27d98yXFdXxxtvvEFENHt3T0RwxRVXMG7cuK3Gz5w5s7C7m3yx2MwsZ+3atey999707NmTZ599lkceeQSAbt268dZbbwHQu3dv1q1bt2WecePGcd1117F+/XoAVq5cySuvvNLsMvr06cNee+3Fgw8+CMDNN9+8VV1XX331lmU999xzbNiwgeOPP54ZM2bQ2NjIyy+/zG9/+9t2W2e3CMzMcsaPH8/06dM58sgjede73sUxxxwDwJQpUzjyyCM56qijuPnmm3n/+9/P8OHDmTBhAtOmTeOZZ57h2GOPBaBXr1785Cc/oa6urtnlXH/99VsuFufP/s8++2yWL1/OUUcdRUTQr18/7rjjDk455RRmz57NEUccwWGHHbalW6k97HC/WTxq1Kjw7xGYWXt55plnOPzwwzs7jHZVbZ0kzY+IUdXKu2vIzKzknAjMzErO1wjK7oqRnR1BbS6Y39kRmO203CIwMys5JwIzs5JzIjAzKzlfIyjICZfP6ewQajK7e+tlzKxz3H333Vx00UU0NjZy9tlnM3Xq1EKW40RgZtaK9j6xm33x2FbLNDY2ct5553HvvfdSX1/P6NGjOemkkxg6dGi7xgLuGjIz65Lmzp3LIYccwkEHHcRuu+3Gaaedxp133lnIspwIzMy6oJUrVzJw4MAtw/X19axcubKQZTkRmJl1QdUe/+Onj5qZlUh9fT0rVqzYMtzQ0ED//v0LWZYTgZlZFzR69GgWL17MsmXLePPNN5kxYwYnnXRSIcvyXUNmZl3QrrvuypVXXsm4ceNobGzkrLPOYtiwYcUsq5Bazcx2IrXc7lmEiRMnMnHixMKX464hM7OScyIwMys5JwIzs5JzIjAzKzknAjOzknMiMDMruUITgaTxkhZJWiJpm+enStpL0i8lPSFpgaQzi4zHzGxHcdZZZ7HffvsxfPjwwpdV2PcIJNUBVwEnAg3APEl3RcTCXLHzgIUR8VFJ/YBFkm6OiDeLisvMrM3a+7e9a/gN7smTJ3P++efz6U9/un2XXUWRLYKjgSURsTQd2GcAJ1eUCaC3sicp9QL+BGwuMCYzsx3C8ccfzz777NMhyyoyEQwAVuSGG9K4vCuBw4GXgKeAiyLi7QJjMjOzCkUmgmrPS618ruo44HGgPzACuFLSnttUJE2R9KikR1999dX2jtPMrNSKTAQNwMDccD3ZmX/emcAvIrMEWAa8u7KiiLgmIkZFxKh+/foVFrCZWRkVmQjmAYdKGiJpN+A04K6KMi8CHwSQtD/wLmBpgTGZmVmFwhJBRGwGzgdmAc8At0bEAknnSjo3FbsUOE7SU8BvgC9HxKqiYjIz21GcfvrpHHvssSxatIj6+nquvfbawpZV6GOoI2ImMLNi3PTc+5eADxcZg5nZO1bD7Z7t7ZZbbumwZfmbxWZmJedEYGZWck4EZmYl50RgZqUXUfkVpx3X9qyLE4GZlVqPHj1YvXr1TpEMIoLVq1fTo0ePNs3nH683s1Krr6+noaGBneWpBT169KC+vr5N8zgRmFmpdevWjSFDhnR2GJ3KXUNmZiXnRGBmVnJOBGZmJedEYGZWck4EZmYl50RgZlZyTgRmZiXnRGBmVnJOBGZmJedEYGZWck4EZmYl50RgZlZyTgRmZiXnRGBmVnJOBGZmJedEYGZWck4EZmYl50RgZlZyTgRmZiXnRGBmVnJOBGZmJedEYGZWck4EZmYl50RgZlZyTgRmZiXnRGBmVnJOBGZmJedEYGZWcoUmAknjJS2StETS1GbKjJX0uKQFku4vMh4zM9vWrkVVLKkOuAo4EWgA5km6KyIW5sr0AX4AjI+IFyXtV1Q8ZmZWXZEtgqOBJRGxNCLeBGYAJ1eU+STwi4h4ESAiXikwHjMzq6LmFoGk44DB+Xki4sYWZhkArMgNNwDvqyhzGNBN0hygN/C9Vuo0M7N2VlMikHQTcDDwONCYRgfQ0kFbVcZFleWPBD4I7A48LOmRiHiuYvlTgCkAgwYNqiVkMzOrUa0tglHA0IioPJC3pAEYmBuuB16qUmZVRGwANkh6AHgPsFUiiIhrgGsARo0a1ZYYzMysFbVeI3ga+Ks21j0POFTSEEm7AacBd1WUuRP4a0m7SupJ1nX0TBuXY2Zm70CtLYK+wEJJc4FNTSMj4qTmZoiIzZLOB2YBdcB1EbFA0rlp+vSIeEbS3cCTwNvAf0bE09u5LmZmth1qTQT/uj2VR8RMYGbFuOkVw9OAadtTv5mZvXM1JYKIuF/S/sDoNGqub/U0M9s51HSNQNLHgbnAqcDHgd9L+liRgZmZWceotWvon4HRTa0ASf2A+4CfFxWYmZl1jFrvGtqloitodRvmNTOzLqzWFsHdkmYBt6ThT1BxEdjMzHZMtV4s/qKkvwPeT/aN4Wsi4vZCIzMzsw5R87OGIuI24LYCYzEzs07QYiKQ9GBEjJG0jq2fEyQgImLPQqMzM7PCtZgIImJM+tu7Y8IxM7OOVuv3CG6qZZyZme14ar0FdFh+QFLT46PNzGwH12IikHRJuj5wpKQ/p9c64I9kTw41M7MdXIuJICL+A9gLuDEi9kyv3hGxb0Rc0jEhmplZkVrtGoqIt8l+LMbMzHZCtV4jeETS6NaLmZnZjqbWL5T9DXCOpBeADfzlewRHFhaZmZl1iFoTwYRCozAzs05TU9dQRLwA9AE+ml590jgzM9vB1fqFsouAm4H90usnki4oMjAzM+sYtXYNfRZ4X0RsAJD0TeBh4IqiAjMzs45R611DAhpzw41pnJmZ7eBqbRFcT/Y7xbeTJYCTgWsLi8rMzDpMrT9M821Jc4AxadSZEfFYYVGZmVmHaevvDovsdwncLWRmtpOo9a6hfwF+DOwN9AWul/SVIgMzM7OOUes1gtOB90bERgBJ3wD+F/h6UYGZmVnHqLVraDnQIzfcHXi+3aMxM7MOV2uLYBOwQNK9ZNcITgQelPR9gIi4sKD4zMysYLUmgtvTq8mc9g/FzMw6Q623j/5Y0m7AYWnUooh4q7iwzMyso9SUCCSNJbtraDnZraMDJX0mIh4oLDIzM+sQtXYNfQv4cEQsApB0GHAL/gF7M7MdXq13DXVrSgIAEfEc0K2YkMzMrCPV2iKYL+la4KY0/ClgfjEhmZlZR6q1RXAusAC4ELgIWJjGtUjSeEmLJC2RNLWFcqMlNUr6WI3xmJlZO2m1RSBpF2B+RAwHvl1rxZLqgKvIvnPQAMyTdFdELKxS7pvArLYEbmZm7aPVFkFEvA08IWlQG+s+GlgSEUsj4k1gBtnjqytdANwGvNLG+s3MrB3Ueo3gALJvFs8FNjSNjIiTWphnALAiN9wAvC9fQNIA4BTgBGB0jbGYmVk7qjURfG076q72qOqoGP4u8OWIaJSaf7K1pCnAFIBBg9raMDEzs5a0mAgk9SC7KHwI8BRwbURsrrHuBmBgbrgeeKmizChgRkoCfYGJkjZHxB35QhFxDXANwKhRoyqTiZmZvQOttQh+DLwF/A8wARhKdtdQLeYBh0oaAqwETgM+mS8QEUOa3ku6AfhVZRIwM7NitZYIhkbEEQDpewRza604IjZLOp/sbqA64LqIWCDp3DR9+nbGbGZm7ai1RLDlwXLpwN6myiNiJjCzYlzVBBARk9tUuZmZtYvWEsF7JP05vRewexoWEBGxZ6HRmZlZ4VpMBBFR11GBmJlZ56j1ERNmZraTciIwMys5JwIzs5JzIjAzKzknAjOzknMiMDMrOScCM7OScyIwMys5JwIzs5JzIjAzKzknAjOzknMiMDMrOScCM7OScyIwMys5JwIzs5JzIjAzKzknAjOzknMiMDMrOScCM7OScyIwMys5JwIzs5JzIjAzKzknAjOzknMiMDMrOScCM7OScyIwMys5JwIzs5JzIjAzKzknAjOzknMiMDMrOScCM7OScyIwMys5JwIzs5IrNBFIGi9pkaQlkqZWmf4pSU+m10OS3lNkPGZmtq3CEoGkOuAqYAIwFDhd0tCKYsuAD0TEkcClwDVFxWNmZtUV2SI4GlgSEUsj4k1gBnByvkBEPBQRr6XBR4D6AuMxM7MqikwEA4AVueGGNK45nwV+XWA8ZmZWxa4F1q0q46JqQelvyBLBmGamTwGmAAwaNKi94jMzM4ptETQAA3PD9cBLlYUkHQn8J3ByRKyuVlFEXBMRoyJiVL9+/QoJ1sysrIpMBPOAQyUNkbQbcBpwV76ApEHAL4AzIuK5AmMxM7NmFNY1FBGbJZ0PzALqgOsiYoGkc9P06cC/APsCP5AEsDkiRhUVk5mZbavIawRExExgZsW46bn3ZwNnFxmDmZm1zN8sNjMrOScCM7OScyIwMys5JwIzs5JzIjAzKzknAjOzknMiMDMrOScCM7OScyIwMys5JwIzs5JzIjAzKzknAjOzknMiMDMruUKfPmpmXc8Jl8/p7BBqMvvisZ0dQmm4RWBmVnJOBGZmJedEYGZWck4EZmYl54vFZtY1XTGysyOozQXzOzuCd8wtAjOzknMiMDMrOScCM7OScyIwMys5JwIzs5JzIjAzKzknAjOzknMiMDMrOScCM7OScyIwMys5JwIzs5JzIjAzKzknAjOzknMiMDMrOScCM7OScyIwMyu5QhOBpPGSFklaImlqlemS9P00/UlJRxUZj5mZbauwRCCpDrgKmAAMBU6XNLSi2ATg0PSaAlxdVDxmZlZdkS2Co4ElEbE0It4EZgAnV5Q5GbgxMo8AfSQdUGBMZmZWocjfLB4ArMgNNwDvq6HMAODlfCFJU8haDADrJS1q31DLS9AXWNXZcbTqQnV2BNbBvG+2uwObm1BkIqi2dWI7yhAR1wDXtEdQtjVJj0bEqM6Ow6yS982OU2TXUAMwMDdcD7y0HWXMzKxARSaCecChkoZI2g04DbirosxdwKfT3UPHAGsj4uXKiszMrDiFdQ1FxGZJ5wOzgDrguohYIOncNH06MBOYCCwBXgfOLCoea5a73Kyr8r7ZQRSxTZe8mZmViL9ZbGZWck4EZmYl50RgZlZyRX6PwLogSe8m+0b3ALLvbLwE3BURz3RqYGbWadwiKBFJXyZ71IeAuWS3+Aq4pdpDAc26Akm+m7BgvmuoRCQ9BwyLiLcqxu8GLIiIQzsnMrPmSXoxIgZ1dhw7M3cNlcvbQH/ghYrxB6RpZp1C0pPNTQL278hYysiJoFy+APxG0mL+8rC/QcAhwPmdFZQZ2cF+HPBaxXgBD3V8OOXiRFAiEXG3pMPIHhE+gOyfrAGYFxGNnRqcld2vgF4R8XjlBElzOjyakvE1AjOzkvNdQ2ZmJedEYGZWck4E1iVIapT0eO41eDvqmFTld7ELIeksSU9JelLS05Iqf4Y1X3aspF+1sf45kt7xj7JI+ldJFzczzRdhDfDFYus63oiIEe+wjklkFx0X1jqDpF0jYnNbFiKpHvhn4KiIWCupF9CvLXW0N0l1bb3gHxHHFRWP7VjcIrAuS9JISfdLmi9plqQD0vjPSZon6QlJt0nqKek44CRgWmpRHJw/q5bUV9Ly9H6ypJ9J+iVwj6Q9JF2X6nyspbP7ZD9gHbAeICLWR8SyVPchku5Lsf2vpIPTPL0k/VzSs5JulqRU/oNpmU+lGLpX2Q5XS3pU0gJJX8uNXy7pXyQ9CJwq6UJJC1MrZUauiqFpWyyVdGFu/vXp71hJD0i6Pc0/XZKPDWUSEX751ekvoBF4PL1uB7qR3T/eL03/BNmPGwHsm5vv68AF6f0NwMdy0+YAo9L7vsDy9H4y2W2z+6Thy4C/T+/7AM8Be5B9+W5mlVjryH5w6UXgeuCjuWm/B05J73sAPYGxwFqyn2LdBXgYGJOmrwAOS+VvBL5QJfZ9csudAxyZhpcDX8ot+yWge9N6pL//mrZj97QNVgPd0rT16e9YYCNwUFrGvfnt6NfO/3LXkHUVW3UNSRoODAfuTSfPdUDTz5gOl/R1soN2L7KDclvdGxF/Su8/DJyU60vvAQyK7EF8EytnjIhGSeOB0cAHge9IGgl8CxgQEbenchvTugDMjYiGNPw4MJisVbEsIp5LVf8YOA/4bsUiPy5pCllX7gHAUKDpm7g/zZV7ErhZ0h3AHbnx/x0Rm4BNkl4h+/JWQ8Uy5kbE0hTfLWSJ6ueV6247JycC66pE9vyjY6tMuwGYFBFPSJpMdkZbzWb+0v3Zo2Lahopl/V1ELKo1uIgIsgf3zZV0L1nL4NstzLIp976R7H9PrS1H0hDgYmB0RLwm6Qa2Xpf8enwEOJ6si+yrkoa1sOxtVqmVYduJuR/QuqpFQD9JxwJI6pY7sPUGXpbUDfhUbp51aVqT5cDI9P5jLSxrFnBBrt/+vS0FJqm/pKNyo0YAL0TEn4EGSZNSue6SerZQ1bPAYEmHpOEzgPsryuxJdrBfK2l/YEIzMe0CDIyI3wJf4i+tpVodLWlIqucTwINtmNd2cE4E1iVFxJtkB+9vSnqC7NpB010uXyXri7+X7GDaZAbwxXTx9WDgcuAf0m2SfVtY3KVk1ySelPR0Gm464M+sUr4bcHm68Ps42YHzojTtDOBCZQ9Rewj4qxbWcSNwJvAzSU+RPfhvekWZJ4DHgAXAdcDvmqmuDvhJqucx4DsRsaaFda70MPAN4GlgGdl1GisJP2LCrOQkjQUujoi/7eRQrJO4RWBmVnJuEZiZlZxbBGZmJedEYGZWck4EZmYl50RgZlZyTgRmZiXnRGBmVnL/H8xugIXlwN/QAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "proportion_attendance('scholarship');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Looks like there is a small, but consistent, difference in the proportion of attendance rate based on receiving scholarship \"Bolsa Familia\". Patients who receive seem less likely to show up, wheres those who do not receive the governmental assistance benefit seem slightly more likely to attend their scheduled appointment.\n",
    "\n",
    "We can use this variable as a possible explanatory variable in our classification model."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Q6) Does any medical condition impact the attendance rate?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ref_list = ['hypertension', 'diabetes', 'alcoholism', 'handicap']\n",
    "\n",
    "proportion_attendance('hypertension');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "proportion_attendance('diabetes');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Both **Hypertension** and **Diabetes** seem to have an impact on attendance rate, where <u>patients with these medical conditions are more likely to show up</u>. We can later use these variables in the classification model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "proportion_attendance('alcoholism');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "proportion_attendance('handicap');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "On the other hand, neither **Alcoholism** nor **Handicap** seem to have <u>any impact over the attendance rate</u>. We should not use these variables to model appointment attendance probability."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Q7) Does receiving the SMS reminder has any impact on the attendance?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "proportion_attendance('sms_received');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Looks like the attendance rate of patients that received the SMS reminder is quite different than those who did not receive. But the interpretation of the observed effect doesn't make sense. It isn't reasonable that a patient would change their mind and decide not to go to their scheduled appointment after receiving the SMS reminder.\n",
    "\n",
    "In this case, even though there is evidence this variable impact the attendance rate, <u>we should ignore it due to lack of explainability</u>."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Q8) Is it possible to, based on the available features, predict whether a patient will show up to their scheduled appointment?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since our response is a binary variable containing data coded as 1 (yes/success) and 0 (no/failure), we can implement the classification algorithm: Logistic Regression. The model predicts the probability (between 0 and 100%) of the dependent variable as a function of the features.\n",
    "\n",
    "> The following procedures were based on the very well written Medium Post from Susan Li, entitled [Building a Logistic Regression in Python, step-by-step](https://towardsdatascience.com/building-a-logistic-regression-in-python-step-by-step-becd4d56c9c8)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Percentage of Attended: 0.80 | Percentage of No-Show: 0.20\n",
      "Ratio: 4:1\n"
     ]
    }
   ],
   "source": [
    "# Checking proportion of Attended: Yes/1 vs No/0\n",
    "prop_attended = df.attended.mean()\n",
    "prop_no_show = 1 - prop_attended\n",
    "ratio = prop_attended / prop_no_show\n",
    "\n",
    "print('Percentage of Attended: {:.2f} | Percentage of No-Show: {:.2f}'\\\n",
    "      .format(prop_attended, prop_no_show))\n",
    "print('Ratio: {:.0f}:1'.format(ratio))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The classes are imbalanced with a ratio of 4:1 of Attended vs No-Show. \n",
    "\n",
    "In order to avoid overfit our model estimation, we need to balance the classes by Over-Sampling the training data (otherwise it would be biased toward predicting Yes/Attended, since it would be right 4 out of 5 times). \n",
    "\n",
    "Let's use the [SMOTE algorithm](https://arxiv.org/pdf/1106.1813.pdf) (Synthetic Minority Oversampling Technique).\n",
    "\n",
    ">As described on the post by Susan Li, this algorithm can be understood, at a high level, as it:\n",
    "+ Works by creating synthetic samples from the minor class (in this case, No-Attended) instead of creating copies.\n",
    "+ Randomly chooses one of the *k-nearest-neighbors* and uses it to create a similar, but randomly tweaked, new observations.\n",
    "\n",
    "To quote Susan's observation:\n",
    "> *\"... by <mark>oversampling only on the training data</mark>, none of the information in the test data is being used to create synthetic observations, therefore, no information will 'bleed' from the test data into the model training.\"*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First, let's create the dummy variables for the features showing potential of being good predictors for out model. We will do so by using a native Pandas function:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# Creating Dummy Variables for categories in Age_Groups\n",
    "dummy = pd.get_dummies(df['age_groups'], prefix='age_gr')\n",
    "df = df.join(dummy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['gender', 'scheduled_day', 'appointment_day', 'age', 'neighborhood',\n",
       "       'scholarship', 'hypertension', 'diabetes', 'alcoholism', 'handicap',\n",
       "       'sms_received', 'attended', 'delta_date', 'delta_weeks', 'delta_days',\n",
       "       'age_groups', 'reg_up_4days', 'reg_more_7days', 'reg_neighb_group1',\n",
       "       'reg_neighb_group2', 'age_gr_(-0.001, 5.0]', 'age_gr_(5.0, 13.0]',\n",
       "       'age_gr_(13.0, 21.0]', 'age_gr_(21.0, 30.0]', 'age_gr_(30.0, 37.0]',\n",
       "       'age_gr_(37.0, 45.0]', 'age_gr_(45.0, 52.0]', 'age_gr_(52.0, 59.0]',\n",
       "       'age_gr_(59.0, 68.0]', 'age_gr_(68.0, 102.0]'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Checking if dummy variables were created correctly\n",
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Keeping only NUMERICAL and potential predictors for model regression\n",
    "to_keep = ['scholarship', 'hypertension', 'diabetes', 'reg_up_4days', 'reg_more_7days',\n",
    "           'reg_neighb_group1','reg_neighb_group2', 'age_gr_(-0.001, 5.0]', 'age_gr_(5.0, 13.0]',\n",
    "           'age_gr_(13.0, 21.0]', 'age_gr_(21.0, 30.0]', 'age_gr_(30.0, 37.0]',\n",
    "           'age_gr_(37.0, 45.0]','age_gr_(45.0, 52.0]', 'age_gr_(52.0, 59.0]',\n",
    "           'age_gr_(59.0, 68.0]', 'age_gr_(68.0, 102.0]']\n",
    "\n",
    "df_model = df[to_keep]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# Balancing response classes with Over-sampling (SMOTE algorithm)\n",
    "from sklearn.model_selection import train_test_split\n",
    "from imblearn.over_sampling import SMOTE\n",
    "\n",
    "X = df_model # features\n",
    "y = df['attended'] # response\n",
    "\n",
    "# randomly splitting a proportion of dataset into train/test subsets\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=0)\n",
    "columns = X_train.columns\n",
    "\n",
    "# generating synthetic samples only for training dataset\n",
    "os = SMOTE(random_state=0)\n",
    "os_data_X, os_data_y = os.fit_resample(X_train, y_train)\n",
    "\n",
    "# storing them back as dataframes\n",
    "X_train = pd.DataFrame(data=os_data_X, columns=columns)\n",
    "y_train = pd.DataFrame(data=os_data_y, columns=['attended'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Length of oversampled data: 123356\n",
      "Number of Attended: 61678\n",
      "Number of No-Show: 61678\n",
      "Percentage of Attended: 0.50 | Percentage of No-Show: 0.50\n",
      "Ratio of Response in oversampled data: 1:1\n"
     ]
    }
   ],
   "source": [
    "# Checking the classes in oversampled data\n",
    "prop_attended = y_train.attended.mean()\n",
    "prop_no_show = 1 - prop_attended\n",
    "ratio = prop_attended / prop_no_show\n",
    "\n",
    "print(\"Length of oversampled data:\",len(X_train))\n",
    "print(\"Number of Attended:\", len(y_train.query('attended == 1')))\n",
    "print(\"Number of No-Show:\", len(y_train.query('attended == 0')))\n",
    "print('Percentage of Attended: {:.2f} | Percentage of No-Show: {:.2f}'\\\n",
    "      .format(prop_attended, prop_no_show) )\n",
    "print('Ratio of Response in oversampled data: {:.0f}:1'.format(ratio))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Synthetic Training Sample | Total rows: 123356 | # Yes-Attended: 61678 | # No-Attended: 61678\n",
      "     Original Sample      | Total rows: 110516 | # Yes-Attended: 88205 | # No-Attended: 22311\n"
     ]
    }
   ],
   "source": [
    "print('Synthetic Training Sample | Total rows: {} | # Yes-Attended: {} | # No-Attended: {}'.\\\n",
    "      format(len(X_train), y_train.attended.sum(), len(y_train) - y_train.attended.sum()))\n",
    "\n",
    "print('     Original Sample      | Total rows: {} | # Yes-Attended: {} | # No-Attended: {}'.\\\n",
    "      format(len(X), y.sum(), len(y) - y.sum()))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now that our training data (shuffled 70% of dataset) has a balanced proportion of the outcome (Attend), we can use it to train our classification model without overfitting it.\n",
    "\n",
    "We can apply a feature selection method to identify the most significant features from our extensive list of possible candidates. \n",
    "\n",
    "But the method described in the post ([**Recursive Feature Elimination**](https://scikit-learn.org/stable/modules/generated/sklearn.feature_selection.RFE.html)) has a limitation: it requires the scientist to arbitrarily set the final number of predictors! Which kind of defeats the purpose of finding the most significant ones.\n",
    "\n",
    "Instead, we are going to implement a variation of such method, the [**Recursive Feature Elimination with Cross Validation**](https://scikit-learn.org/stable/modules/generated/sklearn.feature_selection.RFECV.html#sklearn.feature_selection.RFECV): it recursively build models, identifying the best and worst performing features, setting then aside and continuing the process with the remaining features until they all have been evaluated. It ranks the best performing features, while using Cross-Validation to select the best number of predictors.\n",
    "\n",
    "Let's apply the **RFECV** method using a Logit function as the estimator, removing only 1 feature at a time, and using 5-fold cross-validation:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "# transform y_train into array of shape (n,) - as required to fit the model\n",
    "y_train = y_train.attended"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Applying RFE with Cross Validation\n",
    "from sklearn.feature_selection import RFECV\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "logit = LogisticRegression(solver='liblinear')\n",
    "\n",
    "rfeCV = RFECV(logit, step=1,cv=5, n_jobs=-1)\n",
    "rfeCV = rfeCV.fit(X_train, y_train)\n",
    "# print(rfeCV.support_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "scholarship\n",
      "diabetes\n",
      "reg_up_4days\n",
      "reg_more_7days\n",
      "reg_neighb_group1\n",
      "reg_neighb_group2\n",
      "age_gr_(5.0, 13.0]\n",
      "age_gr_(13.0, 21.0]\n",
      "age_gr_(21.0, 30.0]\n",
      "age_gr_(30.0, 37.0]\n",
      "age_gr_(37.0, 45.0]\n",
      "age_gr_(59.0, 68.0]\n",
      "age_gr_(68.0, 102.0]\n"
     ]
    }
   ],
   "source": [
    "top_predictors = [k for k,v in zip(X_train.columns, rfeCV.support_) if v == True]\n",
    "[print(i) for i in top_predictors]; # 12 or 7 predictors"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's filter the training sample to include only the top performing features identified by the RFECV method:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = X_train[top_predictors]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's fit a Logistic Regression model in order to evaluate the significance of the identified features. We will do so using <code>statsmodels</code> since it has a more detailed result summary:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization terminated successfully.\n",
      "         Current function value: 0.642914\n",
      "         Iterations 5\n",
      "                           Results: Logit\n",
      "=====================================================================\n",
      "Model:               Logit             Pseudo R-squared:  0.072      \n",
      "Dependent Variable:  attended          AIC:               158640.6321\n",
      "Date:                2022-09-06 13:40  BIC:               158767.0288\n",
      "No. Observations:    123356            Log-Likelihood:    -79307.    \n",
      "Df Model:            12                LL-Null:           -85504.    \n",
      "Df Residuals:        123343            LLR p-value:       0.0000     \n",
      "Converged:           1.0000            Scale:             1.0000     \n",
      "No. Iterations:      5.0000                                          \n",
      "---------------------------------------------------------------------\n",
      "                      Coef.  Std.Err.    z     P>|z|   [0.025  0.975]\n",
      "---------------------------------------------------------------------\n",
      "scholarship          -0.1754   0.0198  -8.8422 0.0000 -0.2143 -0.1365\n",
      "diabetes             -0.1571   0.0250  -6.2959 0.0000 -0.2061 -0.1082\n",
      "reg_up_4days          0.8360   0.0124  67.1531 0.0000  0.8116  0.8604\n",
      "reg_more_7days       -0.4350   0.0127 -34.1217 0.0000 -0.4600 -0.4100\n",
      "reg_neighb_group1     0.1151   0.0183   6.2768 0.0000  0.0791  0.1510\n",
      "reg_neighb_group2    -0.2341   0.0230 -10.1565 0.0000 -0.2793 -0.1889\n",
      "age_gr_(5.0, 13.0]   -0.3296   0.0212 -15.5692 0.0000 -0.3710 -0.2881\n",
      "age_gr_(13.0, 21.0]  -0.4974   0.0204 -24.4401 0.0000 -0.5373 -0.4575\n",
      "age_gr_(21.0, 30.0]  -0.4222   0.0195 -21.6670 0.0000 -0.4604 -0.3840\n",
      "age_gr_(30.0, 37.0]  -0.2170   0.0212 -10.2571 0.0000 -0.2584 -0.1755\n",
      "age_gr_(37.0, 45.0]  -0.2345   0.0206 -11.3684 0.0000 -0.2750 -0.1941\n",
      "age_gr_(59.0, 68.0]   0.1013   0.0225   4.5094 0.0000  0.0573  0.1453\n",
      "age_gr_(68.0, 102.0]  0.1294   0.0234   5.5260 0.0000  0.0835  0.1753\n",
      "=====================================================================\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Logistic Regression - Evaluating Features using statsmodels (better result summary)\n",
    "import statsmodels.api as sm\n",
    "\n",
    "logit = sm.Logit(y_train, X_train)\n",
    "result = logit.fit()\n",
    "print(result.summary2())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "All variables presented p-value smaller than the significance level of 5%. This means the **RFECV** did a good job selecting the best performing features.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After the third round, the model looks good. All features have p-value less than significance level of 5%, meaning they are significant.\n",
    "\n",
    "Also, the coefficients have the correct expected signal, as mentioned through the analysis."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Odd's Ratio\n",
    "We can interpret the Odd's ratio by calculating $exp(coef.)$:\n",
    ">**Obs**: it is necessary to correct the Odd's Ratio for values where $exp(coef.) < 1$. By inverting, we transform it into the regular proportion/ratio form, but it is necessary to add a negative sign, since it has an inverted effect."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Coefficient</th>\n",
       "      <th>Odds Ratio</th>\n",
       "      <th>Ranking</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>reg_up_4days</td>\n",
       "      <td>2.3</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>age_gr_(13.0, 21.0]</td>\n",
       "      <td>-1.6</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>reg_more_7days</td>\n",
       "      <td>-1.5</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>age_gr_(21.0, 30.0]</td>\n",
       "      <td>-1.5</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>age_gr_(5.0, 13.0]</td>\n",
       "      <td>-1.4</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>reg_neighb_group2</td>\n",
       "      <td>-1.3</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>age_gr_(37.0, 45.0]</td>\n",
       "      <td>-1.3</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>scholarship</td>\n",
       "      <td>-1.2</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>diabetes</td>\n",
       "      <td>-1.2</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>age_gr_(30.0, 37.0]</td>\n",
       "      <td>-1.2</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>reg_neighb_group1</td>\n",
       "      <td>1.1</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>age_gr_(59.0, 68.0]</td>\n",
       "      <td>1.1</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>age_gr_(68.0, 102.0]</td>\n",
       "      <td>1.1</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Coefficient  Odds Ratio  Ranking\n",
       "2           reg_up_4days         2.3        1\n",
       "7    age_gr_(13.0, 21.0]        -1.6        2\n",
       "3         reg_more_7days        -1.5        3\n",
       "8    age_gr_(21.0, 30.0]        -1.5        3\n",
       "6     age_gr_(5.0, 13.0]        -1.4        4\n",
       "5      reg_neighb_group2        -1.3        5\n",
       "10   age_gr_(37.0, 45.0]        -1.3        5\n",
       "0            scholarship        -1.2        6\n",
       "1               diabetes        -1.2        6\n",
       "9    age_gr_(30.0, 37.0]        -1.2        6\n",
       "4      reg_neighb_group1         1.1        7\n",
       "11   age_gr_(59.0, 68.0]         1.1        7\n",
       "12  age_gr_(68.0, 102.0]         1.1        7"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "coefficients = list(result.params.values)\n",
    "odds_ratio = np.exp(coefficients)\n",
    "\n",
    "# Correcting Odds Ratio for values <1\n",
    "odds_ratio = [i if i >= 1 else -1/i for i in odds_ratio]\n",
    "\n",
    "df_odds_ratio = pd.DataFrame({'Coefficient':top_predictors , 'Odds Ratio': odds_ratio})\n",
    "df_odds_ratio['Odds Ratio'] = df_odds_ratio['Odds Ratio'].round(1)\n",
    "\n",
    "import scipy.stats as ss\n",
    "df_odds_ratio['Ranking'] = ss.rankdata([-1 * abs(i) for i in df_odds_ratio['Odds Ratio']],\n",
    "                                       method='dense')\n",
    "df_odds_ratio.sort_values(by='Ranking')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here is the Odd's Ratio interpretation, ordered by the features with the most impact over the attendance likelihood:\n",
    "\n",
    "1. Appointments scheduled less than 4 days in advance are 2.3 times more likely to attend;\n",
    "2. Patients between the ages of 13 and 21 years old were 1.6 times <u>less likely</u> to show up;\n",
    "3. Tied in third place being 1.5 times less likely to show up:\n",
    "    - <mark>Appointments scheduled more than 7 days in advance</mark>\n",
    "    - <mark>Patients between the ages of 21 and 30 years</mark>\n",
    "4. Patients wwithin the ages of 5 to 13 years were 1.4 times <u>less likely</u> to show up;\n",
    "   - Patients residing in one of the neighborhoods in **Group 2** *(Jesus de Nazareth, Itarare, Santos Dumont)*\n",
    "\n",
    "5. Tied in fifth place being 1.3 times more likely to show up:\n",
    "    - <mark> - Patients residing in one of the neighborhoods in **Group 2** *(Jesus de Nazareth, Itarare, Santos Dumont)*</mark>\n",
    "    - <mark>Patients between age 37 and 45 years </mark>\n",
    "    \n",
    "6. Tied in sixth place, being 1.2 times <u>less likely</u> to show up:\n",
    "    - Patients participating in the governmental assistance program (scholarship: *Bolsa Familia*)\n",
    "    - Patients with diabetes\n",
    "    - Patients between the ages of 30 to 37 years\n",
    "7. Tied in seventh place, being 1.1 times more likely to show up to their scheduled appointment:\n",
    "    - Patients residing in one of the neighborhoods in **Group 1** *(Santa Martha, Jardim da Penha, Vila Rubim)* are 1.1 times more likely  to attend.\n",
    "    - <mark>Patients between 59 and 68 years old</mark>\n",
    "    - <mark>Patients between 68 and 102 years old</mark>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model Performance\n",
    "Ok, now we are ready to evaluate the performance of our Logit Model:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {color: black;background-color: white;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-1 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-1 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>LogisticRegression(solver=&#x27;liblinear&#x27;)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" checked><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">LogisticRegression</label><div class=\"sk-toggleable__content\"><pre>LogisticRegression(solver=&#x27;liblinear&#x27;)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "LogisticRegression(solver='liblinear')"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Logistic Regression - Model Performance\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn import metrics\n",
    "\n",
    "logit = LogisticRegression(solver='liblinear')\n",
    "logit.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Making sure our test sample contains only the significant features previously identified and used in the training sample:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy of Logistic Regression classifier on test set: 0.63\n"
     ]
    }
   ],
   "source": [
    "features = list(X_train.columns.values)\n",
    "X_test = X_test[features]\n",
    "print('Accuracy of Logistic Regression classifier on test set: {:.2f}'\\\n",
    "      .format(logit.score(X_test, y_test)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now let's evaluate **precision** and **recall** based on the *Confusion Matrix*:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[ 4411  2217]\n",
      " [10093 16434]]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "y_pred = logit.predict(X_test)\n",
    "\n",
    "confusion_matrix = confusion_matrix(y_test, y_pred)\n",
    "print(confusion_matrix)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "| Response | Predicted NO | Predicted YES |\n",
    "| --- | --- | --- |\n",
    "| Actual NO | 4,411 | 2,217 |\n",
    "| Actual YES | 10,093 | 16,434 |\n",
    "\n",
    "The result is telling us that we have **4,411 + 16,434 = 20,845** correct predictions (True Positives + True Negatives) and **2,217 + 10,093 = 12,310** incorrect predictions (False Positives + False Negatives)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As described in Scikit Learn:\n",
    "\n",
    "The **precision** is the ratio tp / (tp + fp) where tp is the number of true positives and fp the number of false positives. It can be understood as the proportion of correct predictions over all predictions made for each outcome (Yes/No).\n",
    "\n",
    "The **recall** is the ratio tp / (tp + fn) where tp is the number of true positives and fn the number of false negatives. It can be understood as the proportion of correct predictions over the total number of occurrences of each known outcome (Yes/No).\n",
    "\n",
    "recall is intuitively the ability of the classifier to find all the positive samples.\n",
    "\n",
    "The **F-beta** score can be interpreted as a weighted harmonic mean of the precision and recall, where an F-beta score reaches its best value at 1 and worst score at 0.\n",
    "\n",
    "The F-beta score weights the recall more than the precision by a factor of beta. beta = 1.0 means recall and precision are equally important.\n",
    "\n",
    "The **support** is the number of occurrences of each class in y_test."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.30      0.67      0.42      6628\n",
      "           1       0.88      0.62      0.73     26527\n",
      "\n",
      "    accuracy                           0.63     33155\n",
      "   macro avg       0.59      0.64      0.57     33155\n",
      "weighted avg       0.77      0.63      0.67     33155\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "print(classification_report(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can interpret that our model is quite pessimistic, making several predictions that a patient would NOT show up for their appointment, but only getting it right 30% of the time. By contrast, the model doesn't make as much positive predictions as it should - since the vast majority of outcomes in the testing sample is positive. This suggests the synthetic generated observations in our training sample (in order to balance the classes) actually protected the model from overfitting.\n",
    "\n",
    "Overall, the Logistic model predicts the right outcome 63% of the time ($\\frac{TP + TN}{Total}$)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='conclusions'></a>\n",
    "## Conclusions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After thoroughly investigating all the research questions, we were able to identify that some features seems to actually impact the likelihood of a patient attending the appointment scheduled. To be more specific, some particular groups of those features showed an attendance rate different than the other groups and/or the overall average.\n",
    "> Although, it is important to highlight that, without the proper statistical testing, we cannot ascertain with statistical significance there is indeed a correlation between the features and the attendance rate. \n",
    "\n",
    ">This descriptive statistical analysis allow us to identify features with possible correlation to the dependent variable - which would be interesting to further investigate using hypothesis testing and/or regression models.\n",
    "\n",
    "Here is a summary of the initial findings:\n",
    "+ Gender does not seem to have any impact with the attendance rate;\n",
    "+ Age groups:\n",
    "    - **Less than 5 years and over 45 years old**: seems to be more likely to show up\n",
    "    - **Between 5 and 45 years old**: seems to be less likely to show up\n",
    "+ Date difference between appointment and when it was scheduled:\n",
    "    - **Appointments scheduled up to 4 days in advance**: seems to be more likely to show up\n",
    "    - **Appointments scheduled more than 7 days in advance**: less likely to attend\n",
    "+ Neighborhood groups:\n",
    "    - **Group 1** (Jardim Camburi,  Jardim da Penha, Santa Martha): seems to be more likely to show up\n",
    "    - **Group 2** (Itararé, Jesus de Nazareth, Santos Dumont): seems to be less likely to show up\n",
    "+ Patients receiving Scholarship (Bolsa Familia) are less likely to show up;\n",
    "+ Medical Conditions: patients suffering from alcoholism and diabetes seems to be slightly more likely to attend\n",
    "+ SMS reminder: even though there is evidence receiving an SMS reminder seems to make the patients less likely to show up, we should ignore this variable due to lack of explainability.\n",
    "\n",
    "<hr>\n",
    "After fitting a Logistic Regression model, we ended up with 12 best performing and statistically significant explanatory variables. Their coefficients have the correct expected signal, based on the initial analysis.\n",
    "\n",
    "According to the Odd's ratio analysis, we are able to rank the features based on their impact on the attendance rate:\n",
    "\n",
    "1. Appointments scheduled less than 4 days in advance are 2.3 times more likely to attend;\n",
    "2. Patients between the ages of 13 and 21 years old were 1.6 times <u>less likely</u> to show up;\n",
    "3. Tied in third place being 1.5 times less likely to show up:\n",
    "    - <mark>Appointments scheduled more than 7 days in advance</mark>\n",
    "    - <mark>Patients between the ages of 21 and 30 years</mark>\n",
    "4. Patients wwithin the ages of 5 to 13 years were 1.4 times <u>less likely</u> to show up;\n",
    "   - Patients residing in one of the neighborhoods in **Group 2** *(Jesus de Nazareth, Itarare, Santos Dumont)*\n",
    "\n",
    "5. Tied in fifth place being 1.3 times more likely to show up:\n",
    "    - <mark> - Patients residing in one of the neighborhoods in **Group 2** *(Jesus de Nazareth, Itarare, Santos Dumont)*</mark>\n",
    "    - <mark>Patients between age 37 and 45 years </mark>\n",
    "    \n",
    "6. Tied in sixth place, being 1.2 times <u>less likely</u> to show up:\n",
    "    - Patients participating in the governmental assistance program (scholarship: *Bolsa Familia*)\n",
    "    - Patients with diabetes\n",
    "    - Patients between the ages of 30 to 37 years\n",
    "7. Tied in seventh place, being 1.1 times more likely to show up to their scheduled appointment:\n",
    "    - Patients residing in one of the neighborhoods in **Group 1** *(Santa Martha, Jardim da Penha, Vila Rubim)* are 1.1 times more likely  to attend.\n",
    "    - <mark>Patients between 59 and 68 years old</mark>\n",
    "    - <mark>Patients between 68 and 102 years old</mark>\n",
    "\n",
    "It is possible to see the highlighted variables actually have a similar impact. These age groups could be merged together and the dummy variable would still be able to capture its influence.\n",
    "\n",
    "The final Logistic model is quite pessimistic, making several predictions that a patient would NOT show up for their appointment, but only getting it right 30% of the time. By contrast, the model doesn't make as much positive predictions as it should - since the vast majority of outcomes in the testing sample is positive. \n",
    "\n",
    "Interestingly, this suggests the synthetic generated observations in our training sample (in order to balance the classes) did indeed protect the model from overfitting. This seem to be a good approach for handling datasets with unbalanced classes.\n",
    "\n",
    "Overall, the Logistic model predicts the right outcome 63% of the time ($\\frac{TP + TN}{Total}$). Which is better than flipping a coin, but still has room for improvement before it can be considered useful.\n",
    "\n",
    "### Recommendation for Future Research\n",
    "The advantage of using a Logistic Regression model is it's explanatory power: it allows for interesting inference capabilities such as Odd's Ratio and Elasticity analysis. Although it lacks prediction accuracy power.\n",
    "\n",
    "For this particular case study, the inference power of the Logit model is rendered useless, since the doctors cannot control the characteristics of their patients. Their main goal is to accurately predict whether or not a patient is going to show up for their scheduled appointment.\n",
    "\n",
    "Therefore, it would be interesting to fit the data using more advanced methods, such as Naive Bayes, Support Vector Machine (SVM), Boosting, Random Forest, and even Artificial Neural Network. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.12"
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}