notebooks/random_forest.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import os\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.model_selection import RandomizedSearchCV\n",
    "from sklearn.metrics import mean_squared_error\n",
    "from sklearn.metrics import make_scorer\n",
    "from pandas import DataFrame, concat\n",
    "from pprint import pprint\n",
    "\n",
    "parent_dir = os.path.abspath(os.path.join(os.getcwd(), os.pardir))\n",
    "os.chdir(parent_dir)\n",
    "\n",
    "assert os.path.exists('data') == True, (\n",
    "    'diretório data não encontrado. '\n",
    "    'Certifique-se de que o diretório data existe no diretório raiz desse '\n",
    "    'projeto, com os arquivos necessários dentro.'\n",
    ")\n",
    "\n",
    "os.makedirs('results', exist_ok=True)\n",
    "\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Constantes"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Os valores abaixo serão utilizados para retornar os valores de previsão para a escala original. \n",
    "\n",
    "Além de desnormalizar é necessário readicionar a tendência que foi retirada a partir da diferenciação da série do ICMS-RJ."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Constantes para a validação\n",
    "# 25 valores para criar 13 intervalos de 12 meses\n",
    "icms_initial_values = [3054647241.0, 2899400405.0, 2718750710.0, \n",
    "                       2784291530.0, 2988242758.0, 2728014302.0, \n",
    "                       3400196101.0, 3228952138.0, 2873691468.0, \n",
    "                       2694165362.0, 4479663014.0, 2867019408.0, \n",
    "                       3586300282.0, 3192884893.0, 2763545556.0, \n",
    "                       3193576905.0, 2983787440.0, 2737799049.08, \n",
    "                       2753485166.27, 2931044987.21, 3114528093.4, \n",
    "                       3124850205.99, 3143348373.33, 3490147653.84,\n",
    "                       3859316053.66]\n",
    "\n",
    "cb_initial_values = [443.81, 438.36, 441.19, 440.06, 446.03, 445.58, \n",
    "                     421.89, 417.05, 418.48, 443.69, 460.24, 466.75, \n",
    "                     460.46, 464.47, 496.33, 515.58, 492.93, 498.67, \n",
    "                     479.28, 462.24, 458.21, 462.57, 455.37, 516.91,\n",
    "                     644.0]\n",
    "\n",
    "petr_initial_values = [2724, 2729, 2672, 2714, 2718, 2707, 2690, 2614, \n",
    "                       2578, 2717, 2671, 2794, 2730, 2582, 2655, 2700, \n",
    "                       2834, 2654, 2877, 3097, 3038, 3075, 3203, 3216,\n",
    "                       2969]\n",
    "\n",
    "trib_initial_values = [4869304104.0, 3911455913.0, 3553175853.0, \n",
    "                       3557404009.0, 3632787716.0, 3305049163.0, \n",
    "                       4047070676.0, 3901651824.0, 3444939721.0, \n",
    "                       3272112477.0, 5096900924.0, 3456062224.0, \n",
    "                       5464521206.0, 4185558828.0, 3621893885.0, \n",
    "                       3959255086.0, 3592748799.0, 3300752325.67, \n",
    "                       3321406806.42, 3508106369.08, 3641413728.43, \n",
    "                       3665239552.24, 3657823494.71, 4063564350.02,\n",
    "                       5496860058.37]\n",
    "\n",
    "ener_initial_values = [20078.0, 19690.0, 20853.0, 20733.0, 19736.0, 19179.0, \n",
    "                       19273.0, 19378.0, 19600.0, 20110.0, 20154.0, 20022.0, \n",
    "                       20966.0, 20585.0, 20626.0, 19828.0, 20187.0, 18981.0, \n",
    "                       18700.0, 18900.0, 19292.0, 20171.0, 20623.0, 19593.0,\n",
    "                       21498.0]\n",
    "\n",
    "# Constantes para os testes\n",
    "icms_test_real = [4327920118.33, 3621238631.11, 3349671547.87, \n",
    "                  4123153077.77, 3750000648.34, 4142199873.63, \n",
    "                  3574957480.03, 3409293931.52, 4297149120.6, \n",
    "                  3404457118.32, 3622801505.3, 3334894194.52]\n",
    "\n",
    "initial_icms_test_value = 4021877984.35\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Importando e Plotando as séries\n",
    "No CSV estão salvas apenas as séries selecionadas após a aplicação do DTW e diferenciadas para remover a tendência. Por esse motivo todas as séries são estacionárias."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x1000 with 5 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df = pd.read_csv('data/selected.csv', header=0, names=['ANO', 'MES', 'ICMS', 'CB', 'PETROLEO', 'TRIBUTOS', 'ENERGIA'])\n",
    "plot = df[df.columns[2:]].plot(figsize=(10, 10), subplots=True)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Separando o conjunto de Teste"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(202, 7) (12, 7)\n"
     ]
    }
   ],
   "source": [
    "df_test = df[df['ANO'] == 2022].copy()\n",
    "df = df[df['ANO'] != 2022].copy()\n",
    "print(df.shape, df_test.shape)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Normalizando as séries\n",
    "Os valores Máximo e Mínimo do ICMS são salvos para que seja possível retornar com as previsões para as escala original."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Normalizing data   \n",
    "_min = df['ICMS'].min()\n",
    "_max = df['ICMS'].max()\n",
    "\n",
    "cb_min = df['CB'].min()\n",
    "cb_max = df['CB'].max()\n",
    "\n",
    "petr_min = df['PETROLEO'].min()\n",
    "petr_max = df['PETROLEO'].max()\n",
    "\n",
    "trib_min = df['TRIBUTOS'].min()\n",
    "trib_max = df['TRIBUTOS'].max()\n",
    "\n",
    "ener_min = df['ENERGIA'].min()\n",
    "ener_max = df['ENERGIA'].max()\n",
    "\n",
    "for col in df.columns[2:]:\n",
    "    df[col] = (df[col] - df[col].min()) / (df[col].max() - df[col].min())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>ANO</th>\n",
       "      <th>MES</th>\n",
       "      <th>ICMS</th>\n",
       "      <th>CB</th>\n",
       "      <th>PETROLEO</th>\n",
       "      <th>TRIBUTOS</th>\n",
       "      <th>ENERGIA</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2004.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.453713</td>\n",
       "      <td>0.680444</td>\n",
       "      <td>0.587234</td>\n",
       "      <td>0.426264</td>\n",
       "      <td>0.496627</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2004.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.494609</td>\n",
       "      <td>0.664873</td>\n",
       "      <td>0.457447</td>\n",
       "      <td>0.446347</td>\n",
       "      <td>0.616310</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2004.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.473085</td>\n",
       "      <td>0.730537</td>\n",
       "      <td>0.465957</td>\n",
       "      <td>0.437884</td>\n",
       "      <td>0.266940</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2004.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.502279</td>\n",
       "      <td>0.686920</td>\n",
       "      <td>0.657447</td>\n",
       "      <td>0.480698</td>\n",
       "      <td>0.438838</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2004.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>0.486677</td>\n",
       "      <td>0.612923</td>\n",
       "      <td>0.578723</td>\n",
       "      <td>0.454444</td>\n",
       "      <td>0.498387</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      ANO  MES      ICMS        CB  PETROLEO  TRIBUTOS   ENERGIA\n",
       "0  2004.0  3.0  0.453713  0.680444  0.587234  0.426264  0.496627\n",
       "1  2004.0  4.0  0.494609  0.664873  0.457447  0.446347  0.616310\n",
       "2  2004.0  5.0  0.473085  0.730537  0.465957  0.437884  0.266940\n",
       "3  2004.0  6.0  0.502279  0.686920  0.657447  0.480698  0.438838\n",
       "4  2004.0  7.0  0.486677  0.612923  0.578723  0.454444  0.498387"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Definição de funções"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def series_to_supervised(data, columns, n_in=1, n_out=1, dropnan=True):\n",
    "    \"\"\"\n",
    "    Frame a time series as a supervised learning dataset.\n",
    "    Arguments:\n",
    "    data: Sequence of observations as a list or NumPy array.\n",
    "    n_in: Number of lag observations as input (X).\n",
    "    n_out: Number of observations as output (y).\n",
    "    dropnan: Boolean whether or not to drop rows with NaN values.\n",
    "    Returns:\n",
    "    Pandas DataFrame of series framed for supervised learning.\n",
    "    \"\"\"\n",
    "    n_vars = 1 if type(data) is list else data.shape[1]\n",
    "    df = DataFrame(data)\n",
    "    cols, names = list(), list()\n",
    "    # input sequence (t-n, ... t-1)\n",
    "    for i in range(n_in, 0, -1):\n",
    "        cols.append(df.shift(i))\n",
    "        names += [(f'{columns[j]}(t-{i})') for j in range(n_vars)]\n",
    "    # forecast sequence (t, t+1, ... t+n)\n",
    "    for i in range(0, n_out):\n",
    "        cols.append(df.shift(-i))\n",
    "        if i == 0:\n",
    "            names += [(f'{columns[j]}(t)') for j in range(n_vars)]\n",
    "        else:\n",
    "            names += [(f'{columns[j]}(t+{i})') for j in range(n_vars)]\n",
    "    # put it all together\n",
    "    agg = concat(cols, axis=1)\n",
    "    agg.columns = names\n",
    "    # drop rows with NaN values\n",
    "    if dropnan:\n",
    "        agg.dropna(inplace=True)\n",
    "    return agg\n",
    "\n",
    "def get_best_parameters(best_dict):\n",
    "    \"\"\"\n",
    "    Function to get the best parameters from the dictionary\n",
    "    Parameters:\n",
    "    best_dict: Dictionary returned by the grid search\n",
    "    return: Tuple with the best parameters\n",
    "        ne: n_estimators\n",
    "        mss: min_samples_split\n",
    "        msl: min_samples_leaf\n",
    "        mf: max_features\n",
    "        md: max_depth\n",
    "        b: bootstrap\n",
    "    \"\"\"\n",
    "    ne = best_dict['n_estimators']\n",
    "    mss = best_dict['min_samples_split']\n",
    "    msl = best_dict['min_samples_leaf']\n",
    "    mf = best_dict['max_features']\n",
    "    md = best_dict['max_depth']\n",
    "    b = best_dict['bootstrap']\n",
    "    return ne, mss, msl, mf, md, b\n",
    "\n",
    "def print_mean_error(name, rmse, mae, mape):\n",
    "    \"\"\"\n",
    "    Function to print the mean error\n",
    "    Parameters:\n",
    "    name: Name of the case\n",
    "    rmse: numpy array with the rmse values\n",
    "    mae: numpy array with the mae values\n",
    "    mape: numpy array with the mape values\n",
    "    return: None\n",
    "    \"\"\"\n",
    "    mean_rmse = np.mean(rmse)\n",
    "    mean_mae = np.mean(mae)\n",
    "    mean_mape = np.mean(mape)\n",
    "    print(f'{name}')\n",
    "    print(f'\\tRMSE: {mean_rmse:.2f}')\n",
    "    print(f'\\tMAE: {mean_mae:.2f}')\n",
    "    print(f'\\tMAPE: {mean_mape:.2f}')\n",
    "    print('-'*40)\n",
    "\n",
    "def denorm_and_retrend(y_pred, y_val_case, icms_max, icms_min, \n",
    "                       s_max, s_min, icms_initial, s_initial):\n",
    "    \"\"\"\n",
    "    Function to remove the normalization and retrend the data\n",
    "    Parameters:\n",
    "    y_pred: numpy array with the predicted values\n",
    "    y_val_case: numpy array with the real values of the case (validation)\n",
    "    icms_max: maximum value of the icms\n",
    "    icms_min: minimum value of the icms\n",
    "    s_max: maximum value of the serie of the case\n",
    "    s_min: minimum value of the serie of the case\n",
    "    icms_initial: initial value of the icms for retrend icms\n",
    "    s_initial: initial value of the serie for retrend icms\n",
    "    return: \n",
    "    y_pred: numpy array with the predicted values denormalized and retrended\n",
    "    y_val_case: numpy array with the real values of the case (validation) denormalized and retrended\n",
    "    \"\"\"                       \n",
    "    y_pred = np.array(y_pred).reshape(12, 2)\n",
    "    norm1, norm2 = np.split(y_pred, 2, axis=1)\n",
    "    norm1 = norm1 * (icms_max - icms_min) + icms_min\n",
    "    norm2 = norm2 * (s_max - s_min) + s_min\n",
    "\n",
    "    y_val_case = np.array(y_val_case).reshape(12, 2)\n",
    "    val_norm1, val_norm2 = np.split(y_val_case, 2, axis=1)\n",
    "    val_norm1 = val_norm1 * (icms_max - icms_min) + icms_min\n",
    "    val_norm2 = val_norm2 * (s_max - s_min) + s_min \n",
    "\n",
    "    for j in range(12):\n",
    "        if j == 0:\n",
    "            norm1[j] = norm1[j] + icms_initial\n",
    "            val_norm1[j] = val_norm1[j] + icms_initial\n",
    "            norm2[j] = norm2[j] + s_initial\n",
    "            val_norm2[j] = val_norm2[j] + s_initial\n",
    "        else:\n",
    "            norm1[j] = norm1[j] + norm1[j-1]\n",
    "            val_norm1[j] = val_norm1[j] + val_norm1[j-1]\n",
    "            norm2[j] = norm2[j] + norm2[j-1]\n",
    "            val_norm2[j] = val_norm2[j] + val_norm2[j-1]\n",
    "\n",
    "    y_pred = np.concatenate((norm1, norm2), axis=1).reshape(-1, 1)\n",
    "    y_val_case = np.concatenate((val_norm1, val_norm2), axis=1).reshape(-1, 1)\n",
    "\n",
    "    return y_pred, y_val_case\n",
    "\n",
    "def calculate_error(y_pred, y_val):\n",
    "    \"\"\"\n",
    "    Function to calculate the errors RMSE, MAE and MAPE\n",
    "    Parameters:\n",
    "    y_pred: numpy array with the predicted values (without normalization and retrended)\n",
    "    y_val: numpy array with the real values (without normalization and retrended)\n",
    "    return:\n",
    "    rmse: RMSE value\n",
    "    mae: MAE value\n",
    "    mape: MAPE value\n",
    "    \"\"\"\n",
    "    rmse = np.sqrt(mean_squared_error(y_pred, y_val))\n",
    "    mae = np.mean(np.abs(y_val - y_pred))\n",
    "    mape = np.mean(np.abs((y_val - y_pred) / y_val)) * 100\n",
    "\n",
    "    return rmse, mae, mape\n",
    "\n",
    "def denorm_and_retrend_all(y_pred, y_val, initial, cb_initial, petr_initial, \n",
    "                           trib_initial, ener_initial):\n",
    "    \"\"\"\n",
    "    Function to remove the normalization and retrend the data of the case\n",
    "    that all series are used\n",
    "    Parameters:\n",
    "    y_pred: numpy array with the predicted values\n",
    "    y_val: numpy array with the real values of the case (validation)\n",
    "    initial: initial value of the serie for retrend icms\n",
    "    cb_initial: initial value of the serie for retrend cb\n",
    "    petr_initial: initial value of the serie for retrend petr\n",
    "    trib_initial: initial value of the serie for retrend trib   \n",
    "    ener_initial: initial value of the serie for retrend ener\n",
    "    \"\"\"\n",
    "    y_pred = np.array(y_pred).reshape(12, 5)\n",
    "    norm1, norm2, norm3, norm4, norm5 = np.split(y_pred, 5, axis=1)\n",
    "    norm1 = norm1 * (_max - _min) + _min\n",
    "    norm2 = norm2 * (cb_max - cb_min) + cb_min\n",
    "    norm3 = norm3 * (petr_max - petr_min) + petr_min\n",
    "    norm4 = norm4 * (trib_max - trib_min) + trib_min\n",
    "    norm5 = norm5 * (ener_max - ener_min) + ener_min\n",
    "\n",
    "    y_val = np.array(y_val).reshape(12, 5)\n",
    "    norm1_val, norm2_val, norm3_val, norm4_val, norm5_val = np.split(y_val, 5, axis=1)\n",
    "    norm1_val = norm1_val * (_max - _min) + _min\n",
    "    norm2_val = norm2_val * (cb_max - cb_min) + cb_min\n",
    "    norm3_val = norm3_val * (petr_max - petr_min) + petr_min\n",
    "    norm4_val = norm4_val * (trib_max - trib_min) + trib_min\n",
    "    norm5_val = norm5_val * (ener_max - ener_min) + ener_min\n",
    "\n",
    "    for j in range(12):\n",
    "        if j == 0:\n",
    "            norm1[j] = norm1[j] + initial\n",
    "            norm2[j] = norm2[j] + cb_initial\n",
    "            norm3[j] = norm3[j] + petr_initial\n",
    "            norm4[j] = norm4[j] + trib_initial\n",
    "            norm5[j] = norm5[j] + ener_initial\n",
    "            norm1_val[j] = norm1_val[j] + initial\n",
    "            norm2_val[j] = norm2_val[j] + cb_initial\n",
    "            norm3_val[j] = norm3_val[j] + petr_initial\n",
    "            norm4_val[j] = norm4_val[j] + trib_initial\n",
    "            norm5_val[j] = norm5_val[j] + ener_initial\n",
    "        else:\n",
    "            norm1[j] = norm1[j] + norm1[j-1]\n",
    "            norm2[j] = norm2[j] + norm2[j-1]\n",
    "            norm3[j] = norm3[j] + norm3[j-1]\n",
    "            norm4[j] = norm4[j] + norm4[j-1]\n",
    "            norm5[j] = norm5[j] + norm5[j-1]\n",
    "            norm1_val[j] = norm1_val[j] + norm1_val[j-1]\n",
    "            norm2_val[j] = norm2_val[j] + norm2_val[j-1]\n",
    "            norm3_val[j] = norm3_val[j] + norm3_val[j-1]\n",
    "            norm4_val[j] = norm4_val[j] + norm4_val[j-1]\n",
    "            norm5_val[j] = norm5_val[j] + norm5_val[j-1]\n",
    "\n",
    "    y_pred = np.concatenate((norm1, norm2, norm3, norm4, norm5), axis=1)\n",
    "    y_pred = y_pred.reshape(-1, 1)\n",
    "    y_val = np.concatenate((norm1_val, norm2_val, norm3_val, norm4_val, norm5_val), axis=1)\n",
    "    y_val = y_val.reshape(-1, 1)\n",
    "\n",
    "    return y_pred, y_val\n",
    "\n",
    "def get_prediction(test_values, model, initial):\n",
    "    \"\"\"\n",
    "    Function used in test set to predict the next 12 months in a multistep\n",
    "    forecast. The prediction is made using the model passed as parameter.\n",
    "    Parameters:\n",
    "    test_values: numpy array with the values of the last line of the validation set\n",
    "    model: model used to predict the next 12 months\n",
    "    initial: initial value of the serie for retrend\n",
    "    return:\n",
    "    pred: numpy array with the predicted values denormalized and retrended\n",
    "    \"\"\"\n",
    "    pred = []\n",
    "\n",
    "    for i in range(12):\n",
    "        y_hat = model.predict(test_values.reshape(1, -1))\n",
    "        test_values = np.concatenate([test_values[2:], y_hat.reshape(-1)], axis=0)\n",
    "        #Desnormalizing\n",
    "        y_hat = y_hat * (_max - _min) + _min\n",
    "        pred.append(y_hat[0][0])\n",
    "\n",
    "    #Retrending\n",
    "    for i in range(12):\n",
    "        if i == 0:\n",
    "            pred[i] = pred[i] + initial\n",
    "        else:\n",
    "            pred[i] = pred[i] + pred[i-1]\n",
    "\n",
    "    pred = np.array(pred)\n",
    "    return pred\n",
    "\n",
    "def print_results(name, pred, real):\n",
    "    \"\"\"\n",
    "    Print the results of the model on the test set\n",
    "    Parameters:\n",
    "    name: name of the case\n",
    "    pred: numpy array with the predicted values\n",
    "    real: numpy array with the real values\n",
    "    return: None\n",
    "    \"\"\"\n",
    "    rmse, mae, mape = calculate_error(pred, real)\n",
    "    print(f'{name}')\n",
    "    print(f'\\tRMSE: {rmse:.2f}')\n",
    "    print(f'\\tMAE: {mae:.2f}')\n",
    "    print(f'\\tMAPE: {mape:.2f}')\n",
    "    print('-'*40)\n",
    " "
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Univariado\n",
    "Primeiro vamos usar apenas o ICMS para realizar a previsão multi-step."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(190, 13)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "icms = series_to_supervised(list(df['ICMS']), ['ICMS'], 12, 1)\n",
    "icms.shape\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### GridSearch com CV para a Random Forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 3 folds for each of 100 candidates, totalling 300 fits\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'bootstrap': True,\n",
      " 'max_depth': 10,\n",
      " 'max_features': 'auto',\n",
      " 'min_samples_leaf': 4,\n",
      " 'min_samples_split': 5,\n",
      " 'n_estimators': 1000}\n"
     ]
    }
   ],
   "source": [
    "# Number of trees in random forest\n",
    "n_estimators = [int(x) for x in np.linspace(start = 200, stop = 2000, num = 10)]\n",
    "# Number of features to consider at every split\n",
    "max_features = ['auto', 'sqrt']\n",
    "# Maximum number of levels in tree\n",
    "max_depth = [int(x) for x in np.linspace(10, 210, num = 21)]\n",
    "max_depth.append(None)\n",
    "# Minimum number of samples required to split a node\n",
    "min_samples_split = [2, 5, 10, 20, 30, 50]\n",
    "# Minimum number of samples required at each leaf node\n",
    "min_samples_leaf = [1, 2, 4, 10, 30, 50]\n",
    "# Method of selecting samples for training each tree\n",
    "bootstrap = [True, False]\n",
    "# Create the random grid\n",
    "random_grid = {'n_estimators': n_estimators,\n",
    "               'max_features': max_features,\n",
    "               'max_depth': max_depth,\n",
    "               'min_samples_split': min_samples_split,\n",
    "               'min_samples_leaf': min_samples_leaf,\n",
    "               'bootstrap': bootstrap}\n",
    "\n",
    "# Use the random grid to search for best hyperparameters\n",
    "# First create the base model to tune\n",
    "rf = RandomForestRegressor()\n",
    "# Random search of parameters, using 3 fold cross validation, \n",
    "# search across 100 different combinations, and use all available cores\n",
    "rf_random = RandomizedSearchCV(estimator = rf, param_distributions = random_grid, \n",
    "                               n_iter = 100, cv = 3, scoring='neg_mean_squared_error',\n",
    "                               verbose=2, n_jobs = -1)\n",
    "# Fit the random search model\n",
    "rf_random.fit(icms.values[:, :-1], icms.values[:, -1])\n",
    "\n",
    "ne = rf_random.best_params_['n_estimators']\n",
    "mss = rf_random.best_params_['min_samples_split']\n",
    "msl = rf_random.best_params_['min_samples_leaf']\n",
    "mf = rf_random.best_params_['max_features']\n",
    "md = rf_random.best_params_['max_depth']\n",
    "b = rf_random.best_params_['bootstrap']\n",
    "pprint(rf_random.best_params_)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Rodando uma Random Forest univariada com os valores padrões\n",
    "\n",
    "Para os 36 meses de validação, foi calculado 25 casos de uma previsão Multistep de 12 meses. A métrica de erro foi obtida através da média do erro desses 25 casos.\n",
    "\n",
    "São mostrados 12 gráficos que correspondem aos intervalos [0, 12], [2, 14], [4, 16] ... [24, 36]."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Divisão em Treino e Validação\n",
    "\n",
    "Dos 190 registros existentes após o janelamento, os últimos 36 (2018,2019, 2021) (~20%) foram escolhidos para fazer parte do conjunto de Validação, enquanto os outros fazem parte do conjunto de treino."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Dividindo em treino e validação\n",
    "train = icms.values[:154, :]\n",
    "val = icms.values[154:, :]\n",
    "\n",
    "#Dividindo em X e y\n",
    "X_train, y_train = train[:, :-1], train[:, -1]\n",
    "X_val, y_val = val[:, :-1], val[:, -1]"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Treinamento com parâmetro padrão e 100 estimators\n",
    "\n",
    "Também são calculado as métricas de erro no conjunto de validação"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 400x200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Random Forest com parâmetros padrões e 100 estimadores\t\n",
    "plot_many = False\n",
    "plot_one = True\n",
    "df_rf = RandomForestRegressor(n_estimators=100)\n",
    "df_rf.fit(X_train, y_train) # (154, 12), (154,)\n",
    "\n",
    "\n",
    "default_rmse = []\n",
    "default_mae = []\n",
    "default_mape = []\n",
    "\n",
    "if plot_many:\n",
    "    fig, axs = plt.subplots(13, figsize=(4, 20))\n",
    "if plot_one:\n",
    "    chosen = np.random.randint(0, 25)\n",
    "    fig, axs = plt.subplots(1, figsize=(4, 2))\n",
    "\n",
    "for i in range(25):\n",
    "\n",
    "    initial = icms_initial_values[i]\n",
    "    y_pred = []\n",
    "\n",
    "    X_case = X_val[i:i+1].copy()\n",
    "    \n",
    "\n",
    "    # Multi-step prediction\n",
    "    for j in range(12):\n",
    "        yhat = df_rf.predict(X_case)\n",
    "        y_pred.append(yhat)\n",
    "        X_case = np.concatenate([X_case[:, 1:], yhat.reshape(1, -1)], axis=1)\n",
    "    \n",
    "    # Desnormalizing\n",
    "    y_pred = np.array(y_pred).reshape(-1, 1)\n",
    "    y_pred = y_pred * (_max - _min) + _min\n",
    "    y_val_case = y_val[i:i+12].reshape(-1, 1) * (_max - _min) + _min\n",
    "\n",
    "    # Retrending\n",
    "    for j in range(12):\n",
    "        if j == 0:\n",
    "            y_pred[j] = y_pred[j] + initial\n",
    "            y_val_case[j] = y_val_case[j] + initial\n",
    "        else:\n",
    "            y_pred[j] = y_pred[j] + y_pred[j-1]\n",
    "            y_val_case[j] = y_val_case[j] + y_val_case[j-1]\n",
    "\n",
    "    #Plotando\n",
    "    if plot_many:\n",
    "        if i % 2 == 0:\n",
    "            axs[i//2].plot(y_val_case, label='Valor real')\n",
    "            axs[i//2].plot(y_pred, label='Valor predito')\n",
    "            axs[i//2].legend()\n",
    "            axs[i//2].set(xticks=np.arange(0, 12, 1), xticklabels=np.arange(i+1, i+13, 1))\n",
    "            axs[i//2].set_title(f\"{i}º até {i+12}° mês\")\n",
    "    elif plot_one:\n",
    "        if i == chosen:\n",
    "            axs.plot(y_val_case, label='Valor real')\n",
    "            axs.plot(y_pred, label='Valor predito')\n",
    "            axs.legend()\n",
    "            axs.set(xticks=np.arange(0, 12, 1), xticklabels=np.arange(i+1, i+13, 1))\n",
    "            axs.set_title(f\"{i}º até {i+12}° mês\")\n",
    "\n",
    "    # Calculating the error\n",
    "    default_rmse.append(np.sqrt(mean_squared_error(y_val_case, y_pred)))\n",
    "    default_mae.append(np.mean(np.abs(y_val_case - y_pred)))\n",
    "    default_mape.append(np.mean(np.abs((y_val_case[:12] - y_pred[:12]) / y_val_case[:12])) * 100)\n",
    "    \n",
    "#Plotando metade dos gráficos de validação criados\n",
    "plt.tight_layout()\n",
    "plt.show()\n",
    "\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Treinamento com parâmetros otimizados pelo grid\n",
    "\n",
    "Também são calculadas as métricas de erro no conjunto de validação"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 400x200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Random Forest\n",
    "plot_many = False\n",
    "plot_one = True\n",
    "best_rf = RandomForestRegressor(n_estimators= ne, min_samples_split = mss,\n",
    "                           min_samples_leaf= msl, max_features= mf,\n",
    "                           max_depth= md, bootstrap= b)\n",
    "best_rf.fit(X_train, y_train) # (154, 12), (154,)\n",
    "\n",
    "best_rmse = []\n",
    "best_mae = []\n",
    "best_mape = []\n",
    "\n",
    "if plot_many:\n",
    "    fig, axs = plt.subplots(13, figsize=(4, 20))\n",
    "elif plot_one:\n",
    "    chosen = np.random.randint(0, 25)\n",
    "    fig, axs = plt.subplots(1, figsize=(4, 2))\n",
    "    \n",
    "for i in range(25):\n",
    "    initial = icms_initial_values[i]\n",
    "    y_pred = []\n",
    "\n",
    "    X_case = X_val[i:i+1].copy()\n",
    "\n",
    "    # Multi-step prediction\n",
    "    for j in range(12):\n",
    "        yhat = best_rf.predict(X_case)\n",
    "        y_pred.append(yhat)\n",
    "        X_case = np.concatenate([X_case[:, 1:], yhat.reshape(1, -1)], axis=1)\n",
    "    \n",
    "    # Desnormalizing\n",
    "    y_pred = np.array(y_pred).reshape(-1, 1)\n",
    "    y_pred = y_pred * (_max - _min) + _min\n",
    "    y_val_case = y_val[i:i+12].reshape(-1, 1) * (_max - _min) + _min\n",
    "\n",
    "    # Retrending\n",
    "    for j in range(12):\n",
    "        if j == 0:\n",
    "            y_pred[j] = y_pred[j] + initial\n",
    "            y_val_case[j] = y_val_case[j] + initial\n",
    "        else:\n",
    "            y_pred[j] = y_pred[j] + y_pred[j-1]\n",
    "            y_val_case[j] = y_val_case[j] + y_val_case[j-1]\n",
    "\n",
    "        #Plotando\n",
    "    if plot_many:\n",
    "        if i % 2 == 0:\n",
    "            axs[i//2].plot(y_val_case, label='Valor real')\n",
    "            axs[i//2].plot(y_pred, label='Valor predito')\n",
    "            axs[i//2].legend()\n",
    "            axs[i//2].set(xticks=np.arange(0, 12, 1), xticklabels=np.arange(i+1, i+13, 1))\n",
    "            axs[i//2].set_title(f\"{i}º até {i+12}° mês\")\n",
    "    elif plot_one:\n",
    "        if i == chosen:\n",
    "            axs.plot(y_val_case, label='Valor real')\n",
    "            axs.plot(y_pred, label='Valor predito')\n",
    "            axs.legend()\n",
    "            axs.set(xticks=np.arange(0, 12, 1), xticklabels=np.arange(i+1, i+13, 1))\n",
    "            axs.set_title(f\"{i}º até {i+12}° mês\")\n",
    "\n",
    "    # Calculating the error\n",
    "    best_rmse.append(np.sqrt(mean_squared_error(y_val_case, y_pred)))\n",
    "    best_mae.append(np.mean(np.abs(y_val_case - y_pred)))\n",
    "    best_mape.append(np.mean(np.abs((y_val_case[:12] - y_pred[:12]) / y_val_case[:12])) * 100)\n",
    "    \n",
    "\n",
    "#Plotando\n",
    "plt.tight_layout()\n",
    "plt.show()\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Mostrando os parâmetros utilizado no caso Padrão e no Otimizado"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Default parameters:\n",
      "\n",
      "{'bootstrap': True,\n",
      " 'ccp_alpha': 0.0,\n",
      " 'criterion': 'squared_error',\n",
      " 'max_depth': None,\n",
      " 'max_features': 'auto',\n",
      " 'max_leaf_nodes': None,\n",
      " 'max_samples': None,\n",
      " 'min_impurity_decrease': 0.0,\n",
      " 'min_samples_leaf': 1,\n",
      " 'min_samples_split': 2,\n",
      " 'min_weight_fraction_leaf': 0.0,\n",
      " 'n_estimators': 100,\n",
      " 'n_jobs': None,\n",
      " 'oob_score': False,\n",
      " 'random_state': None,\n",
      " 'verbose': 0,\n",
      " 'warm_start': False}\n"
     ]
    }
   ],
   "source": [
    "print('Default parameters:\\n')\n",
    "pprint(df_rf.get_params())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Best parameters:\n",
      "\n",
      "{'bootstrap': True,\n",
      " 'ccp_alpha': 0.0,\n",
      " 'criterion': 'squared_error',\n",
      " 'max_depth': 10,\n",
      " 'max_features': 'auto',\n",
      " 'max_leaf_nodes': None,\n",
      " 'max_samples': None,\n",
      " 'min_impurity_decrease': 0.0,\n",
      " 'min_samples_leaf': 4,\n",
      " 'min_samples_split': 5,\n",
      " 'min_weight_fraction_leaf': 0.0,\n",
      " 'n_estimators': 1000,\n",
      " 'n_jobs': None,\n",
      " 'oob_score': False,\n",
      " 'random_state': None,\n",
      " 'verbose': 0,\n",
      " 'warm_start': False}\n"
     ]
    }
   ],
   "source": [
    "print('\\nBest parameters:\\n')\n",
    "pprint(best_rf.get_params())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Validação: Parâmetros Default Univariado\n",
      "\tRMSE: 539828719.18\n",
      "\tMAE: 450468579.59\n",
      "\tMAPE: 15.63\n",
      "----------------------------------------\n",
      "Validação: Parâmetros Otimizados Univariado\n",
      "\tRMSE: 567249277.57\n",
      "\tMAE: 474073016.60\n",
      "\tMAPE: 16.21\n",
      "----------------------------------------\n"
     ]
    }
   ],
   "source": [
    "print_mean_error(\"Validação: Parâmetros Default Univariado\",\n",
    "                 default_rmse, default_mae, default_mape)\n",
    "print_mean_error(\"Validação: Parâmetros Otimizados Univariado\",\n",
    "                best_rmse, best_mae, best_mape)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Multivariado"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Casos multivariados com ICMS e uma série exógena"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Preparando as tabelas com os lags"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "#['ANO', 'MES', 'ICMS', 'CB', 'PETROLEO', 'TRIBUTOS', 'ENERGIA']\n",
    "icms_cb_raw = df[['ICMS', 'CB']].copy()\n",
    "icms_petr_raw = df[['ICMS', 'PETROLEO']].copy() \n",
    "icms_trib_raw = df[['ICMS', 'TRIBUTOS']].copy()\n",
    "icms_ener_raw = df[['ICMS', 'ENERGIA']].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(190, 26)\n",
      "(190, 26)\n",
      "(190, 26)\n",
      "(190, 26)\n"
     ]
    }
   ],
   "source": [
    "icms_cb = series_to_supervised(icms_cb_raw, ['ICMS', 'CB'], 12, 1)\n",
    "icms_petr = series_to_supervised(icms_petr_raw, ['ICMS', 'PETROLEO'], 12, 1)  \n",
    "icms_trib = series_to_supervised(icms_trib_raw, ['ICMS', 'TRIBUTOS'], 12, 1)\n",
    "icms_ener = series_to_supervised(icms_ener_raw, ['ICMS', 'ENERGIA'], 12, 1)\n",
    "print(icms_cb.shape)\n",
    "print(icms_petr.shape)\n",
    "print(icms_trib.shape)\n",
    "print(icms_ener.shape)  "
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### GridSearch CV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 3 folds for each of 100 candidates, totalling 300 fits\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 3 folds for each of 100 candidates, totalling 300 fits\n",
      "Fitting 3 folds for each of 100 candidates, totalling 300 fits\n",
      "Fitting 3 folds for each of 100 candidates, totalling 300 fits\n",
      "{'bootstrap': True,\n",
      " 'max_depth': 40,\n",
      " 'max_features': 'auto',\n",
      " 'min_samples_leaf': 1,\n",
      " 'min_samples_split': 20,\n",
      " 'n_estimators': 200}\n",
      "{'bootstrap': True,\n",
      " 'max_depth': 90,\n",
      " 'max_features': 'auto',\n",
      " 'min_samples_leaf': 50,\n",
      " 'min_samples_split': 10,\n",
      " 'n_estimators': 2000}\n",
      "{'bootstrap': True,\n",
      " 'max_depth': 90,\n",
      " 'max_features': 'auto',\n",
      " 'min_samples_leaf': 1,\n",
      " 'min_samples_split': 5,\n",
      " 'n_estimators': 1600}\n",
      "{'bootstrap': True,\n",
      " 'max_depth': 30,\n",
      " 'max_features': 'auto',\n",
      " 'min_samples_leaf': 4,\n",
      " 'min_samples_split': 50,\n",
      " 'n_estimators': 400}\n"
     ]
    }
   ],
   "source": [
    "# Use the random grid to search for best hyperparameters\n",
    "# First create the base model to tune\n",
    "rf = RandomForestRegressor()\n",
    "# Random search of parameters, using 3 fold cross validation, \n",
    "# search across 100 different combinations, and use all available cores\n",
    "icms_cb_random = RandomizedSearchCV(estimator = rf, param_distributions = random_grid, \n",
    "                                    n_iter = 100, cv = 3, scoring='neg_mean_squared_error',\n",
    "                                    verbose=2, n_jobs = -1)\n",
    "icms_petr_random = RandomizedSearchCV(estimator = rf, param_distributions = random_grid, \n",
    "                                    n_iter = 100, cv = 3, scoring='neg_mean_squared_error',\n",
    "                                    verbose=2, n_jobs = -1)\n",
    "icms_trib_random = RandomizedSearchCV(estimator = rf, param_distributions = random_grid, \n",
    "                                    n_iter = 100, cv = 3, scoring='neg_mean_squared_error',\n",
    "                                    verbose=2, n_jobs = -1)\n",
    "icms_ener_random = RandomizedSearchCV(estimator = rf, param_distributions = random_grid, \n",
    "                                    n_iter = 100, cv = 3, scoring='neg_mean_squared_error',\n",
    "                                    verbose=2, n_jobs = -1)\n",
    "\n",
    "\n",
    "# Fit the random search model\n",
    "icms_cb_random.fit(icms_cb.values[:, :-2], icms_cb.values[:, -2])\n",
    "icms_petr_random.fit(icms_petr.values[:, :-2], icms_petr.values[:, -2])\n",
    "icms_trib_random.fit(icms_trib.values[:, :-2], icms_trib.values[:, -2])\n",
    "icms_ener_random.fit(icms_ener.values[:, :-2], icms_ener.values[:, -2])\n",
    "\n",
    "cb_ne, cb_mss, cb_msl, cb_mf, cb_md, cb_b = get_best_parameters(icms_cb_random.best_params_)\n",
    "petr_ne, petr_mss, petr_msl, petr_mf, petr_md, petr_b = get_best_parameters(icms_petr_random.best_params_)\n",
    "trib_ne, trib_mss, trib_msl, trib_mf, trib_md, trib_b = get_best_parameters(icms_trib_random.best_params_)\n",
    "ener_ne, ener_mss, ener_msl, ener_mf, ener_md, ener_b = get_best_parameters(icms_ener_random.best_params_)\n",
    "\n",
    "pprint(icms_cb_random.best_params_)\n",
    "pprint(icms_petr_random.best_params_)\n",
    "pprint(icms_trib_random.best_params_)\n",
    "pprint(icms_ener_random.best_params_)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Divisão de treinamento e validação"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Cesta Básicas + ICMS\n",
    "#Dividindo em treino e validação\n",
    "cb_train = icms_cb.values[:154, :]\n",
    "cb_val = icms_cb.values[154:, :]    \n",
    "\n",
    "#Dividindo em X e y\n",
    "cb_X_train, cb_y_train = cb_train[:, :-2], cb_train[:, -2:]\n",
    "cb_X_val, cb_y_val = cb_val[:, :-2], cb_val[:, -2:]\n",
    "\n",
    "#Petróleo + ICMS\n",
    "#Dividindo em treino e validação\n",
    "petr_train = icms_petr.values[:154, :]\n",
    "petr_val = icms_petr.values[154:, :]\n",
    "\n",
    "#Dividindo em X e y\n",
    "petr_X_train, petr_y_train = petr_train[:, :-2], petr_train[:, -2:]\n",
    "petr_X_val, petr_y_val = petr_val[:, :-2], petr_val[:, -2:]\n",
    "\n",
    "#Total de Tributos + ICMS\n",
    "#Dividindo em treino e validação\n",
    "trib_train = icms_trib.values[:154, :]\n",
    "trib_val = icms_trib.values[154:, :]\n",
    "\n",
    "#Dividindo em X e y\n",
    "trib_X_train, trib_y_train = trib_train[:, :-2], trib_train[:, -2:]\n",
    "trib_X_val, trib_y_val = trib_val[:, :-2], trib_val[:, -2:]\n",
    "\n",
    "#Energia + ICMS\n",
    "#Dividindo em treino e validação\n",
    "ener_train = icms_ener.values[:154, :]\n",
    "ener_val = icms_ener.values[154:, :]\n",
    "\n",
    "#Dividindo em X e y\n",
    "ener_X_train, ener_y_train = ener_train[:, :-2], ener_train[:, -2:]\n",
    "ener_X_val, ener_y_val = ener_val[:, :-2], ener_val[:, -2:]"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Treinamento dos 4 modelos\n",
    "\n",
    "Caso ICMS + Cesta Básica\n",
    "\n",
    "Caso ICMS + Produção Total de Petróleo \n",
    "\n",
    "Caso ICMS + Total Receita Tributária RJ\n",
    "\n",
    "Caso ICMS + Consumo de Energia Total Sudeste"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor()"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Random Forest\n",
    "cb_rf = RandomForestRegressor(n_estimators=100)\n",
    "cb_rf.fit(cb_X_train, cb_y_train) # (154, 24), (154,)\n",
    "\n",
    "petr_rf = RandomForestRegressor(n_estimators=100)\n",
    "petr_rf.fit(petr_X_train, petr_y_train) # (154, 24), (154,)\n",
    "\n",
    "trib_rf = RandomForestRegressor(n_estimators=100)\n",
    "trib_rf.fit(trib_X_train, trib_y_train) # (154, 24), (154,)\n",
    "\n",
    "ener_rf = RandomForestRegressor(n_estimators=100)\n",
    "ener_rf.fit(ener_X_train, ener_y_train) # (154, 24), (154,)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Treinamento dos 4 modelos com parâmetros otimizados"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor(max_depth=30, min_samples_leaf=4, min_samples_split=50,\n",
       "                      n_estimators=400)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Random Forest\n",
    "opt_cb_rf = RandomForestRegressor(n_estimators=cb_ne, min_samples_split=cb_mss,\n",
    "                                  min_samples_leaf=cb_msl, max_features=cb_mf,\n",
    "                                  max_depth=cb_md, bootstrap=cb_b)\n",
    "opt_cb_rf.fit(cb_X_train, cb_y_train) # (154, 24), (154,)\n",
    "\n",
    "opt_petr_rf = RandomForestRegressor(n_estimators=petr_ne, min_samples_split=petr_mss,\n",
    "                                    min_samples_leaf=petr_msl, max_features=petr_mf,\n",
    "                                    max_depth=petr_md, bootstrap=petr_b)\n",
    "opt_petr_rf.fit(petr_X_train, petr_y_train) # (154, 24), (154,)\n",
    "\n",
    "opt_trib_rf = RandomForestRegressor(n_estimators=trib_ne, min_samples_split=trib_mss,\n",
    "                                    min_samples_leaf=trib_msl, max_features=trib_mf,\n",
    "                                    max_depth=trib_md, bootstrap=trib_b )\n",
    "opt_trib_rf.fit(trib_X_train, trib_y_train) # (154, 24), (154,)\n",
    "\n",
    "opt_ener_rf = RandomForestRegressor(n_estimators=ener_ne, min_samples_split=ener_mss,\n",
    "                                    min_samples_leaf=ener_msl, max_features=ener_mf,\n",
    "                                    max_depth=ener_md, bootstrap=ener_b)\n",
    "opt_ener_rf.fit(ener_X_train, ener_y_train) # (154, 24), (154,)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Cálculo das métricas de erro no conjunto de validação"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "#fig, axs = plt.subplots(13, figsize=(4, 20))\n",
    "#To use all Validation Set with 12 steps ahead horizon\n",
    "# Criando 25 previsões com horizonte de 12 meses\n",
    "# Para os 36 dados do conjunto de validação\n",
    "for i in range(25):\n",
    "\n",
    "    initial = icms_initial_values[i]\n",
    "    cb_initial = cb_initial_values[i]\n",
    "    petr_initial = petr_initial_values[i]\n",
    "    trib_initial = trib_initial_values[i]\n",
    "    ener_initial = ener_initial_values[i]\n",
    "\n",
    "    cb_pred = []\n",
    "    petr_pred = []\n",
    "    trib_pred = []\n",
    "    ener_pred = []\n",
    "    opt_cb_pred = []\n",
    "    opt_petr_pred = []\n",
    "    opt_trib_pred = []\n",
    "    opt_ener_pred = []\n",
    "\n",
    "    cb_case = cb_X_val[i:i+1].copy() # (1,24)\n",
    "    petr_case = petr_X_val[i:i+1].copy() # (1,24)\n",
    "    trib_case = trib_X_val[i:i+1].copy() # (1,24)\n",
    "    ener_case = ener_X_val[i:i+1].copy() # (1,24) \n",
    "    opt_cb_case = cb_X_val[i:i+1].copy() # (1,24)\n",
    "    opt_petr_case = petr_X_val[i:i+1].copy() # (1,24)\n",
    "    opt_trib_case = trib_X_val[i:i+1].copy() # (1,24)\n",
    "    opt_ener_case = ener_X_val[i:i+1].copy() # (1,24)\n",
    "\n",
    "    # Multi-step prediction for each case\n",
    "    for j in range(12):\n",
    "        cb_yhat = cb_rf.predict(cb_case)   \n",
    "        cb_pred.append(cb_yhat)\n",
    "        cb_case = np.concatenate([cb_case[:, 2:], cb_yhat.reshape(1, -1)], axis=1)\n",
    "\n",
    "        petr_yhat = petr_rf.predict(petr_case)\n",
    "        petr_pred.append(petr_yhat)\n",
    "        petr_case = np.concatenate([petr_case[:, 2:], petr_yhat.reshape(1, -1)], axis=1)\n",
    "\n",
    "        trib_yhat = trib_rf.predict(trib_case)\n",
    "        trib_pred.append(trib_yhat)\n",
    "        trib_case = np.concatenate([trib_case[:, 2:], trib_yhat.reshape(1, -1)], axis=1)\n",
    "\n",
    "        ener_yhat = ener_rf.predict(ener_case)\n",
    "        ener_pred.append(ener_yhat)\n",
    "        ener_case = np.concatenate([ener_case[:, 2:], ener_yhat.reshape(1, -1)], axis=1)\n",
    "\n",
    "        opt_cb_yhat = opt_cb_rf.predict(opt_cb_case)\n",
    "        opt_cb_pred.append(opt_cb_yhat)\n",
    "        opt_cb_case = np.concatenate([opt_cb_case[:, 2:], opt_cb_yhat.reshape(1, -1)], axis=1)\n",
    "\n",
    "        opt_petr_yhat = opt_petr_rf.predict(opt_petr_case)\n",
    "        opt_petr_pred.append(opt_petr_yhat)\n",
    "        opt_petr_case = np.concatenate([opt_petr_case[:, 2:], opt_petr_yhat.reshape(1, -1)], axis=1)\n",
    "\n",
    "        opt_trib_yhat = opt_trib_rf.predict(opt_trib_case)\n",
    "        opt_trib_pred.append(opt_trib_yhat)\n",
    "        opt_trib_case = np.concatenate([opt_trib_case[:, 2:], opt_trib_yhat.reshape(1, -1)], axis=1)\n",
    "\n",
    "        opt_ener_yhat = opt_ener_rf.predict(opt_ener_case)\n",
    "        opt_ener_pred.append(opt_ener_yhat)\n",
    "        opt_ener_case = np.concatenate([opt_ener_case[:, 2:], opt_ener_yhat.reshape(1, -1)], axis=1)\n",
    "\n",
    "    \n",
    "\n",
    "    # Desnormalizing and Retrending\n",
    "    cb_pred, cb_val_case = denorm_and_retrend(cb_pred, cb_y_val[i:i+12], \n",
    "                                              _max, _min, cb_max, cb_min, \n",
    "                                              initial, cb_initial)\n",
    "    petr_pred, petr_val_case = denorm_and_retrend(petr_pred, petr_y_val[i:i+12],\n",
    "                                                  _max, _min, petr_max, petr_min,\n",
    "                                                  initial, petr_initial)\n",
    "    trib_pred, trib_val_case = denorm_and_retrend(trib_pred, trib_y_val[i:i+12],\n",
    "                                                  _max, _min, trib_max, trib_min,\n",
    "                                                  initial, trib_initial)\n",
    "    ener_pred, ener_val_case = denorm_and_retrend(ener_pred, ener_y_val[i:i+12],\n",
    "                                                  _max, _min, ener_max, ener_min,\n",
    "                                                  initial, ener_initial)\n",
    "    opt_cb_pred, opt_cb_val_case = denorm_and_retrend(opt_cb_pred, cb_y_val[i:i+12],\n",
    "                                                        _max, _min, cb_max, cb_min,\n",
    "                                                        initial, cb_initial)\n",
    "    opt_petr_pred, opt_petr_val_case = denorm_and_retrend(opt_petr_pred, petr_y_val[i:i+12],\n",
    "                                                            _max, _min, petr_max, petr_min,\n",
    "                                                            initial, petr_initial)\n",
    "    opt_trib_pred, opt_trib_val_case = denorm_and_retrend(opt_trib_pred, trib_y_val[i:i+12],\n",
    "                                                            _max, _min, trib_max, trib_min,\n",
    "                                                            initial, trib_initial)\n",
    "    opt_ener_pred, opt_ener_val_case = denorm_and_retrend(opt_ener_pred, ener_y_val[i:i+12],\n",
    "                                                            _max, _min, ener_max, ener_min,\n",
    "                                                            initial, ener_initial)\n",
    "          \n",
    "    cb_rmse, cb_mae, cb_mape = calculate_error(cb_pred, cb_val_case)      \n",
    "    petr_rmse, petr_mae, petr_mape = calculate_error(petr_pred, petr_val_case)\n",
    "    trib_rmse, trib_mae, trib_mape = calculate_error(trib_pred, trib_val_case)\n",
    "    ener_rmse, ener_mae, ener_mape = calculate_error(ener_pred, ener_val_case)\n",
    "    opt_cb_rmse, opt_cb_mae, opt_cb_mape = calculate_error(opt_cb_pred, opt_cb_val_case)\n",
    "    opt_petr_rmse, opt_petr_mae, opt_petr_mape = calculate_error(opt_petr_pred, opt_petr_val_case)\n",
    "    opt_trib_rmse, opt_trib_mae, opt_trib_mape = calculate_error(opt_trib_pred, opt_trib_val_case)\n",
    "    opt_ener_rmse, opt_ener_mae, opt_ener_mape = calculate_error(opt_ener_pred, opt_ener_val_case)\n",
    "    \n",
    "#Plotando\n",
    "# plt.tight_layout()\n",
    "# plt.show()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Multivariado com todas as séries"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Criação da tabela contendo todas as séries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(190, 65)\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>ICMS(t-12)</th>\n",
       "      <th>CB(t-12)</th>\n",
       "      <th>PETROLEO(t-12)</th>\n",
       "      <th>TRIBUTOS(t-12)</th>\n",
       "      <th>ENERGIA(t-12)</th>\n",
       "      <th>ICMS(t-11)</th>\n",
       "      <th>CB(t-11)</th>\n",
       "      <th>PETROLEO(t-11)</th>\n",
       "      <th>TRIBUTOS(t-11)</th>\n",
       "      <th>ENERGIA(t-11)</th>\n",
       "      <th>...</th>\n",
       "      <th>ICMS(t-1)</th>\n",
       "      <th>CB(t-1)</th>\n",
       "      <th>PETROLEO(t-1)</th>\n",
       "      <th>TRIBUTOS(t-1)</th>\n",
       "      <th>ENERGIA(t-1)</th>\n",
       "      <th>ICMS(t)</th>\n",
       "      <th>CB(t)</th>\n",
       "      <th>PETROLEO(t)</th>\n",
       "      <th>TRIBUTOS(t)</th>\n",
       "      <th>ENERGIA(t)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>0.453713</td>\n",
       "      <td>0.680444</td>\n",
       "      <td>0.587234</td>\n",
       "      <td>0.426264</td>\n",
       "      <td>0.496627</td>\n",
       "      <td>0.494609</td>\n",
       "      <td>0.664873</td>\n",
       "      <td>0.457447</td>\n",
       "      <td>0.446347</td>\n",
       "      <td>0.616310</td>\n",
       "      <td>...</td>\n",
       "      <td>0.413514</td>\n",
       "      <td>0.699110</td>\n",
       "      <td>0.521277</td>\n",
       "      <td>0.383802</td>\n",
       "      <td>0.335289</td>\n",
       "      <td>0.468697</td>\n",
       "      <td>0.635779</td>\n",
       "      <td>0.623404</td>\n",
       "      <td>0.437047</td>\n",
       "      <td>0.546788</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.494609</td>\n",
       "      <td>0.664873</td>\n",
       "      <td>0.457447</td>\n",
       "      <td>0.446347</td>\n",
       "      <td>0.616310</td>\n",
       "      <td>0.473085</td>\n",
       "      <td>0.730537</td>\n",
       "      <td>0.465957</td>\n",
       "      <td>0.437884</td>\n",
       "      <td>0.266940</td>\n",
       "      <td>...</td>\n",
       "      <td>0.468697</td>\n",
       "      <td>0.635779</td>\n",
       "      <td>0.623404</td>\n",
       "      <td>0.437047</td>\n",
       "      <td>0.546788</td>\n",
       "      <td>0.494049</td>\n",
       "      <td>0.701062</td>\n",
       "      <td>0.797872</td>\n",
       "      <td>0.448239</td>\n",
       "      <td>0.672045</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0.473085</td>\n",
       "      <td>0.730537</td>\n",
       "      <td>0.465957</td>\n",
       "      <td>0.437884</td>\n",
       "      <td>0.266940</td>\n",
       "      <td>0.502279</td>\n",
       "      <td>0.686920</td>\n",
       "      <td>0.657447</td>\n",
       "      <td>0.480698</td>\n",
       "      <td>0.438838</td>\n",
       "      <td>...</td>\n",
       "      <td>0.494049</td>\n",
       "      <td>0.701062</td>\n",
       "      <td>0.797872</td>\n",
       "      <td>0.448239</td>\n",
       "      <td>0.672045</td>\n",
       "      <td>0.460907</td>\n",
       "      <td>0.703252</td>\n",
       "      <td>0.568085</td>\n",
       "      <td>0.430126</td>\n",
       "      <td>0.229393</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>0.502279</td>\n",
       "      <td>0.686920</td>\n",
       "      <td>0.657447</td>\n",
       "      <td>0.480698</td>\n",
       "      <td>0.438838</td>\n",
       "      <td>0.486677</td>\n",
       "      <td>0.612923</td>\n",
       "      <td>0.578723</td>\n",
       "      <td>0.454444</td>\n",
       "      <td>0.498387</td>\n",
       "      <td>...</td>\n",
       "      <td>0.460907</td>\n",
       "      <td>0.703252</td>\n",
       "      <td>0.568085</td>\n",
       "      <td>0.430126</td>\n",
       "      <td>0.229393</td>\n",
       "      <td>0.441648</td>\n",
       "      <td>0.590924</td>\n",
       "      <td>0.572340</td>\n",
       "      <td>0.412021</td>\n",
       "      <td>0.456439</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0.486677</td>\n",
       "      <td>0.612923</td>\n",
       "      <td>0.578723</td>\n",
       "      <td>0.454444</td>\n",
       "      <td>0.498387</td>\n",
       "      <td>0.480029</td>\n",
       "      <td>0.733013</td>\n",
       "      <td>0.497872</td>\n",
       "      <td>0.452790</td>\n",
       "      <td>0.469346</td>\n",
       "      <td>...</td>\n",
       "      <td>0.441648</td>\n",
       "      <td>0.590924</td>\n",
       "      <td>0.572340</td>\n",
       "      <td>0.412021</td>\n",
       "      <td>0.456439</td>\n",
       "      <td>0.496061</td>\n",
       "      <td>0.691253</td>\n",
       "      <td>0.497872</td>\n",
       "      <td>0.468748</td>\n",
       "      <td>0.404517</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 65 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    ICMS(t-12)  CB(t-12)  PETROLEO(t-12)  TRIBUTOS(t-12)  ENERGIA(t-12)  \\\n",
       "12    0.453713  0.680444        0.587234        0.426264       0.496627   \n",
       "13    0.494609  0.664873        0.457447        0.446347       0.616310   \n",
       "14    0.473085  0.730537        0.465957        0.437884       0.266940   \n",
       "15    0.502279  0.686920        0.657447        0.480698       0.438838   \n",
       "16    0.486677  0.612923        0.578723        0.454444       0.498387   \n",
       "\n",
       "    ICMS(t-11)  CB(t-11)  PETROLEO(t-11)  TRIBUTOS(t-11)  ENERGIA(t-11)  ...  \\\n",
       "12    0.494609  0.664873        0.457447        0.446347       0.616310  ...   \n",
       "13    0.473085  0.730537        0.465957        0.437884       0.266940  ...   \n",
       "14    0.502279  0.686920        0.657447        0.480698       0.438838  ...   \n",
       "15    0.486677  0.612923        0.578723        0.454444       0.498387  ...   \n",
       "16    0.480029  0.733013        0.497872        0.452790       0.469346  ...   \n",
       "\n",
       "    ICMS(t-1)   CB(t-1)  PETROLEO(t-1)  TRIBUTOS(t-1)  ENERGIA(t-1)   ICMS(t)  \\\n",
       "12   0.413514  0.699110       0.521277       0.383802      0.335289  0.468697   \n",
       "13   0.468697  0.635779       0.623404       0.437047      0.546788  0.494049   \n",
       "14   0.494049  0.701062       0.797872       0.448239      0.672045  0.460907   \n",
       "15   0.460907  0.703252       0.568085       0.430126      0.229393  0.441648   \n",
       "16   0.441648  0.590924       0.572340       0.412021      0.456439  0.496061   \n",
       "\n",
       "       CB(t)  PETROLEO(t)  TRIBUTOS(t)  ENERGIA(t)  \n",
       "12  0.635779     0.623404     0.437047    0.546788  \n",
       "13  0.701062     0.797872     0.448239    0.672045  \n",
       "14  0.703252     0.568085     0.430126    0.229393  \n",
       "15  0.590924     0.572340     0.412021    0.456439  \n",
       "16  0.691253     0.497872     0.468748    0.404517  \n",
       "\n",
       "[5 rows x 65 columns]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_series_raw = df[['ICMS', 'CB', 'PETROLEO', 'TRIBUTOS', 'ENERGIA']]\n",
    "all_df = series_to_supervised(all_series_raw, ['ICMS', 'CB', 'PETROLEO', 'TRIBUTOS', 'ENERGIA'], 12, 1)\n",
    "print(all_df.shape)\n",
    "all_df.head()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Divisão em treino e validação"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Dividindo em treino e validação\n",
    "all_train = all_df.values[:154, :]\n",
    "all_val = all_df.values[154:, :]    \n",
    "\n",
    "#Dividindo em X e y\n",
    "all_X_train, all_y_train = all_train[:, :-5], all_train[:, -5:]\n",
    "all_X_val, all_y_val = all_val[:, :-5], all_val[:, -5:]"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Treinamento e cálculo das métricas de erro no conjunto de validação"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 400x200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Random Forest\n",
    "plot_many = False\n",
    "plot_one = True\n",
    "all_rf = RandomForestRegressor(n_estimators=100)\n",
    "all_rf.fit(all_X_train, all_y_train) # (154, 12), (154,)\n",
    "\n",
    "\n",
    "all_rmse = []\n",
    "all_mae = []\n",
    "all_mape = []\n",
    "\n",
    "if plot_many:\n",
    "    fig, axs = plt.subplots(13, figsize=(4, 20))\n",
    "elif plot_one:\n",
    "    chosen = np.random.randint(0, 25)\n",
    "    fig, axs = plt.subplots(1, figsize=(4, 2))\n",
    "\n",
    "for i in range(25):\n",
    "\n",
    "    initial = icms_initial_values[i]\n",
    "    cb_initial = cb_initial_values[i]\n",
    "    petr_initial = petr_initial_values[i]\n",
    "    trib_initial = trib_initial_values[i]\n",
    "    ener_initial = ener_initial_values[i]\n",
    "    y_pred = []\n",
    "\n",
    "    X_case = all_X_val[i:i+1].copy()\n",
    "\n",
    "    # Multi-step prediction\n",
    "    for j in range(12):\n",
    "        yhat = all_rf.predict(X_case)\n",
    "        y_pred.append(yhat)\n",
    "        X_case = np.concatenate([X_case[:, 5:], yhat.reshape(1, -1)], axis=1)\n",
    "    \n",
    "\n",
    "    y_pred, y_val_case = denorm_and_retrend_all(y_pred, all_y_val[i:i+12], initial,\n",
    "                                                cb_initial, petr_initial, \n",
    "                                                trib_initial, ener_initial)\n",
    "\n",
    "    icms_pred = y_pred[::5]\n",
    "    icms_val_case = y_val_case[::5]\n",
    "\n",
    "    #Plotando\n",
    "    if plot_many:\n",
    "        if i % 2 == 0:\n",
    "            axs[i//2].plot(icms_val_case, label='Valor real')\n",
    "            axs[i//2].plot(icms_pred, label='Valor predito')\n",
    "            axs[i//2].legend()\n",
    "            axs[i//2].set(xticks=np.arange(0, 12, 1), xticklabels=np.arange(i+1, i+13, 1))\n",
    "            axs[i//2].set_title(f\"{i}º até {i+12}° mês\")\n",
    "    elif plot_one:\n",
    "        if i == chosen:\n",
    "            axs.plot(icms_val_case, label='Valor real')\n",
    "            axs.plot(icms_pred, label='Valor predito')\n",
    "            axs.legend()\n",
    "            axs.set(xticks=np.arange(0, 12, 1), xticklabels=np.arange(i+1, i+13, 1))\n",
    "            axs.set_title(f\"{i}º até {i+12}° mês\")\n",
    "\n",
    "    # Calculating the error\n",
    "    all_rmse.append(np.sqrt(mean_squared_error(y_val_case, y_pred)))\n",
    "    all_mae.append(np.mean(np.abs(y_val_case - y_pred)))\n",
    "    all_mape.append(np.mean(np.abs((y_val_case - y_pred) / y_val_case)) * 100)\n",
    "    \n",
    "#Plotando\n",
    "plt.tight_layout()\n",
    "plt.show()\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Comparação de todos os resultados no conjunto de validação"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Validação: Parâmetros Default Univariado\n",
      "\tRMSE: 539828719.18\n",
      "\tMAE: 450468579.59\n",
      "\tMAPE: 15.63\n",
      "----------------------------------------\n",
      "Validação: Parâmetros Otimizados Univariado\n",
      "\tRMSE: 567249277.57\n",
      "\tMAE: 474073016.60\n",
      "\tMAPE: 16.21\n",
      "----------------------------------------\n",
      "Validação: CB + ICMS Multivariado\n",
      "\tRMSE: 492877073.70\n",
      "\tMAE: 319646212.49\n",
      "\tMAPE: 10.45\n",
      "----------------------------------------\n",
      "Validação: CB + ICMS Otimizado Multivariado\n",
      "\tRMSE: 439232598.44\n",
      "\tMAE: 280176785.55\n",
      "\tMAPE: 9.78\n",
      "----------------------------------------\n",
      "Validação: Petróleo + ICMS Multivariado\n",
      "\tRMSE: 399423739.11\n",
      "\tMAE: 246072505.80\n",
      "\tMAPE: 10.57\n",
      "----------------------------------------\n",
      "Validação: Petróleo + ICMS Otimizado Multivariado\n",
      "\tRMSE: 324906187.88\n",
      "\tMAE: 193868675.16\n",
      "\tMAPE: 9.32\n",
      "----------------------------------------\n",
      "Validação: Tributos + ICMS Multivariado\n",
      "\tRMSE: 602782186.27\n",
      "\tMAE: 529734733.52\n",
      "\tMAPE: 9.79\n",
      "----------------------------------------\n",
      "Validação: Tributos + ICMS Otimizado Multivariado\n",
      "\tRMSE: 657856553.27\n",
      "\tMAE: 587171867.11\n",
      "\tMAPE: 10.80\n",
      "----------------------------------------\n",
      "Validação: Energia + ICMS Multivariado\n",
      "\tRMSE: 338294178.85\n",
      "\tMAE: 196935846.18\n",
      "\tMAPE: 6.30\n",
      "----------------------------------------\n",
      "Validação: Energia + ICMS Otimizado Multivariado\n",
      "\tRMSE: 344523041.90\n",
      "\tMAE: 204890947.05\n",
      "\tMAPE: 7.30\n",
      "----------------------------------------\n",
      "Validação: Todos os parâmetros Multivariado\n",
      "\tRMSE: 428574748.20\n",
      "\tMAE: 217972625.05\n",
      "\tMAPE: 9.43\n",
      "----------------------------------------\n"
     ]
    }
   ],
   "source": [
    "#Comparando os resultados\n",
    "print_mean_error(\"Validação: Parâmetros Default Univariado\",\n",
    "                 default_rmse, default_mae, default_mape)\n",
    "print_mean_error(\"Validação: Parâmetros Otimizados Univariado\",\n",
    "                best_rmse, best_mae, best_mape)\n",
    "print_mean_error(\"Validação: CB + ICMS Multivariado\",\n",
    "                cb_rmse, cb_mae, cb_mape)\n",
    "print_mean_error(\"Validação: CB + ICMS Otimizado Multivariado\",\n",
    "                 opt_cb_rmse, opt_cb_mae, opt_cb_mape)\n",
    "print_mean_error(\"Validação: Petróleo + ICMS Multivariado\",\n",
    "                petr_rmse, petr_mae, petr_mape)\n",
    "print_mean_error(\"Validação: Petróleo + ICMS Otimizado Multivariado\",\n",
    "                    opt_petr_rmse, opt_petr_mae, opt_petr_mape) \n",
    "print_mean_error(\"Validação: Tributos + ICMS Multivariado\",\n",
    "                trib_rmse, trib_mae, trib_mape)\n",
    "print_mean_error(\"Validação: Tributos + ICMS Otimizado Multivariado\",\n",
    "                opt_trib_rmse, opt_trib_mae, opt_trib_mape)\n",
    "print_mean_error(\"Validação: Energia + ICMS Multivariado\",\n",
    "                ener_rmse, ener_mae, ener_mape)\n",
    "print_mean_error(\"Validação: Energia + ICMS Otimizado Multivariado\",\n",
    "                opt_ener_rmse, opt_ener_mae, opt_ener_mape)\n",
    "print_mean_error(\"Validação: Todos os parâmetros Multivariado\",\n",
    "                all_rmse, all_mae, all_mape)      "
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Resultados de Teste"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Teste Univariado"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Using test set Univariate\n",
    "\n",
    "#Getting 12 months before the test set\n",
    "test = icms[-1:].values[0][1:]\n",
    "pred = []\n",
    "\n",
    "for i in range(12):\n",
    "    y_hat = df_rf.predict(test.reshape(1, -1))\n",
    "    test = np.concatenate([test[1:], y_hat], axis=0)    \n",
    "    #Desnormalizing\n",
    "    y_hat = y_hat * (_max - _min) + _min\n",
    "    pred.append(y_hat[0])\n",
    "\n",
    "#Retrending\n",
    "for i in range(12):\n",
    "    if i == 0:\n",
    "        pred[i] = pred[i] + initial_icms_test_value\n",
    "    else:\n",
    "        pred[i] = pred[i] + pred[i-1]\n",
    "\n",
    "pred = np.array(pred)\n",
    "\n",
    "rmse, mae, mape = calculate_error(pred, icms_test_real)\n",
    "\n",
    "plt.plot(icms_test_real, label='Real')\n",
    "plt.plot(pred, label='Predicted')\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test RMSE: 678551913.74\n",
      "Test MAE: 564159525.80\n",
      "Test MAPE: 16.05\n"
     ]
    }
   ],
   "source": [
    "print(f'Test RMSE: {rmse:.2f}')\n",
    "print(f'Test MAE: {mae:.2f}')\n",
    "print(f'Test MAPE: {mape:.2f}')"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Teste Multivariado (ICMS + 1 série Exógena)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Multivariado\n",
    "\n",
    "cb_test = icms_cb[-1:].values[0][2:]\n",
    "petr_test = icms_petr[-1:].values[0][2:]\n",
    "trib_test = icms_trib[-1:].values[0][2:]\n",
    "ener_test = icms_ener[-1:].values[0][2:]\n",
    "\n",
    "cb_pred = get_prediction(cb_test, cb_rf, initial_icms_test_value)\n",
    "petr_pred = get_prediction(petr_test, petr_rf, initial_icms_test_value)\n",
    "trib_pred = get_prediction(trib_test, trib_rf, initial_icms_test_value)\n",
    "ener_pred = get_prediction(ener_test, ener_rf, initial_icms_test_value)\n",
    "opt_cb_pred = get_prediction(cb_test, opt_cb_rf, initial_icms_test_value)\n",
    "opt_petr_pred = get_prediction(petr_test, opt_petr_rf, initial_icms_test_value)\n",
    "opt_trib_pred = get_prediction(trib_test, opt_trib_rf, initial_icms_test_value)\n",
    "opt_ener_pred = get_prediction(ener_test, opt_ener_rf, initial_icms_test_value)\n",
    "\n",
    "plt.figure(figsize=(6,4))\n",
    "plt.plot(icms_test_real, label='Real')\n",
    "plt.plot(cb_pred, label='CB + ICMS Predicted')\n",
    "plt.plot(petr_pred, label='Petróleo + ICMS Predicted')\n",
    "plt.plot(trib_pred, label='Tributos + ICMS Predicted')\n",
    "plt.plot(ener_pred, label='Energia + ICMS Predicted')\n",
    "plt.legend()\n",
    "plt.show()\n",
    "\n",
    "plt.figure(figsize=(6,4))\n",
    "plt.plot(icms_test_real, label='Real')\n",
    "plt.plot(opt_cb_pred, label='CB + ICMS Predicted Otimizado')\n",
    "plt.plot(opt_petr_pred, label='Petróleo + ICMS Predicted Otimizado')\n",
    "plt.plot(opt_trib_pred, label='Tributos + ICMS Predicted Otimizado')\n",
    "plt.plot(opt_ener_pred, label='Energia + ICMS Predicted Otimizado')\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "all_test = all_df[-1:].values[0][5:]\n",
    "\n",
    "all_pred = []\n",
    "\n",
    "for i in range(12):\n",
    "    y_hat = all_rf.predict(all_test.reshape(1, -1))\n",
    "    all_test = np.concatenate([all_test[5:], y_hat.reshape(-1)], axis=0)\n",
    "    #Desnormalizing\n",
    "    y_hat = y_hat * (_max - _min) + _min\n",
    "    all_pred.append(y_hat[0][0])\n",
    "\n",
    "#Retrending\n",
    "for i in range(12):\n",
    "    if i == 0:\n",
    "        all_pred[i] = all_pred[i] + initial_icms_test_value\n",
    "    else:\n",
    "        all_pred[i] = all_pred[i] + all_pred[i-1]\n",
    "\n",
    "all_pred = np.array(all_pred)\n",
    "\n",
    "plt.figure(figsize=(6,4))\n",
    "plt.plot(icms_test_real, label='Real')\n",
    "plt.plot(all_pred, label='Todos as Séries Predicted')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Testes: CB + ICMS Multivariado \n",
      "\tRMSE: 617037924.29\n",
      "\tMAE: 510051386.51\n",
      "\tMAPE: 14.55\n",
      "----------------------------------------\n",
      "Testes: CB + ICMS Multivariado Otimizado\n",
      "\tRMSE: 649067921.31\n",
      "\tMAE: 543378971.84\n",
      "\tMAPE: 15.47\n",
      "----------------------------------------\n",
      "Testes: Petróleo + ICMS Multivariado \n",
      "\tRMSE: 719228050.38\n",
      "\tMAE: 620285591.70\n",
      "\tMAPE: 17.57\n",
      "----------------------------------------\n",
      "Testes: Petróleo + ICMS Multivariado Otimizado\n",
      "\tRMSE: 532869717.75\n",
      "\tMAE: 455561622.43\n",
      "\tMAPE: 12.91\n",
      "----------------------------------------\n",
      "Testes: Tributos + ICMS Multivariado \n",
      "\tRMSE: 545990244.83\n",
      "\tMAE: 460651327.75\n",
      "\tMAPE: 13.08\n",
      "----------------------------------------\n",
      "Testes: Tributos + ICMS Multivariado Otimizado\n",
      "\tRMSE: 501625143.86\n",
      "\tMAE: 431091234.19\n",
      "\tMAPE: 12.17\n",
      "----------------------------------------\n",
      "Testes: Energia + ICMS Multivariado \n",
      "\tRMSE: 507636139.90\n",
      "\tMAE: 447825957.99\n",
      "\tMAPE: 12.62\n",
      "----------------------------------------\n",
      "Testes: Energia + ICMS Multivariado Otimizado\n",
      "\tRMSE: 526608925.89\n",
      "\tMAE: 453151392.35\n",
      "\tMAPE: 12.82\n",
      "----------------------------------------\n",
      "Testes: Todos os parâmetros Multivariado \n",
      "\tRMSE: 608555873.15\n",
      "\tMAE: 500909517.82\n",
      "\tMAPE: 14.31\n",
      "----------------------------------------\n"
     ]
    }
   ],
   "source": [
    "print_results('Testes: ICMS Univariado', pred, icms_test_real)\n",
    "print_results('Testes: CB + ICMS Multivariado ', cb_pred, icms_test_real)\n",
    "print_results('Testes: CB + ICMS Multivariado Otimizado', opt_cb_pred, icms_test_real)\n",
    "print_results('Testes: Petróleo + ICMS Multivariado ', petr_pred, icms_test_real)\n",
    "print_results('Testes: Petróleo + ICMS Multivariado Otimizado', opt_petr_pred, icms_test_real)\n",
    "print_results('Testes: Tributos + ICMS Multivariado ', trib_pred, icms_test_real)  \n",
    "print_results('Testes: Tributos + ICMS Multivariado Otimizado', opt_trib_pred, icms_test_real)\n",
    "print_results('Testes: Energia + ICMS Multivariado ', ener_pred, icms_test_real)\n",
    "print_results('Testes: Energia + ICMS Multivariado Otimizado', opt_ener_pred, icms_test_real)\n",
    "print_results('Testes: Todos os parâmetros Multivariado ', all_pred, icms_test_real)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "icms",
   "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": "3.9.13"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}