Add files via upload

salesplansuccess/api.py

@@ -0,0 +1,117 @@
+'''
+MIT License
+
+Copyright (c) 2022 Alexander Yuryatin
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
+'''
+
+import numpy as np
+import pandas as pd
+import matplotlib.pyplot as plt
+import statsmodels.api as sm
+from scipy import stats
+from datetime import date
+
+class SalesPlanSuccess:
+    def __init__(self, data:pd.DataFrame, plan:int, sample_size:int = 50000):
+        if not isinstance(data, pd.DataFrame):
+            raise TypeError("The parameter 'data' of class SalesPlanSuccess can accept only pandas.DataFrame")
+        if not isinstance(plan, int) and not isinstance(plan, float):
+            raise TypeError("The parameter 'plan' of class SalesPlanSuccess can accept only regular Python integers and floats")
+        if not isinstance(sample_size, int):
+            raise TypeError("The parameter 'sample_size' of class SalesPlanSuccess can accept only regular Python integers")
+        self.tt = data.copy()
+        self.plan = plan
+        self.sample_size = sample_size
+        if (self.sample_size < 1000) or (self.sample_size > 10000000):
+            raise ValueError("The parameter 'sample_size' of class SalesPlanSuccess must be between 1000 and 10 000 000")
+        if self.plan <= 0:
+            raise ValueError("The parameter 'plan' of class SalesPlanSuccess must be a positive number")
+        if set(self.tt.columns) != set(['Year', 'Month', 'Sales']):
+            raise ValueError("The parameter 'data' of class SalesPlanSuccess can only be pandas.DataFrame with three columns with names 'Year', 'Month', and 'Sales'")
+        if self.tt.Year.dtypes != 'int64':
+            raise ValueError("The column 'Year' in parameter 'data' of class SalesPlanSuccess must be of the dtype 'int64'")
+        if self.tt.Month.dtypes != 'int64':
+            raise ValueError("The column 'Month' in parameter 'data' of class SalesPlanSuccess must be of the dtype 'int64'")
+        if self.tt.Sales.dtypes != 'int64' and self.tt.Sales.dtypes != 'float64':
+            raise ValueError("The column 'Sales' in parameter 'data' of Class SalesPlanSuccess must be either of the dtype 'float64' or dtype 'int64'")
+        if (self.tt.Year < 2000).sum() or (self.tt.Year > date.today().year).sum():
+            raise ValueError("The column 'Year' in parameter 'data' of Class SalesPlanSuccess is expected to contain only the years between 2000 and the current year")
+        if (self.tt.Month < 1).sum() or (self.tt.Month > 12).sum():
+            raise ValueError("The column 'Month' in parameter 'data' of Class SalesPlanSuccess is expected to contain only the month numbers between 1 and 12")
+        if (self.tt.Sales <= 0).sum():
+            raise ValueError("The column 'Sales' in parameter 'data' of Class SalesPlanSuccess must contain only positive (non-zero) numbers")
+        if self.tt.isnull().values.any():
+            raise ValueError("The pandas.DataFrame in parameter 'data' of Class SalesPlanSuccess must not contain any empty values")
+        self.tt['qEnd'] = 0
+        self.tt.loc[np.int64(self.tt.Month.values) % 3 == 0, 'qEnd'] = 1
+        self.tt['Sales'] = np.log(self.tt.Sales)
+        self.finalSales = self.tt.Sales.iloc[-1]
+        self.tt['Sales'] = self.tt.Sales.diff()
+        self.tt.dropna(inplace=True)
+        self.tt.reset_index(drop=True, inplace=True)
+        self.model = sm.tsa.ARIMA(endog=self.tt.Sales.values, exog=self.tt.qEnd.values, order=(2, 0, 0))
+        self.finalTwo = self.tt.Sales.iloc[-2:].values.reshape((2,1))
+
+    def fit(self) -> None:
+        self.model_fit = self.model.fit()
+        self.qEnd = (((np.arange(int(self.tt.Month.iloc[-1])+1, 13) % 3) == 0) * self.model_fit.params[1]).reshape((5,1))
+        self.ARs = self.model_fit.params[3:1:-1].reshape((1,2))
+        self.monthsToForecast = 12-int(self.tt.Month.iloc[-1])
+        self.m1 = np.dot(self.ARs, self.finalTwo)
+
+    def simulate(self) -> None:
+        if not hasattr(self, 'model_fit'):
+            raise ValueError("Before calling method 'simulate' of an object of class SalesPlanSuccess you first have to fit the ARIMA model by calling method 'fit' of the same object")
+        self.simul = np.random.normal(loc=self.model_fit.params[0], scale=np.sqrt(self.model_fit.params[4]), size=[self.monthsToForecast, self.sample_size])
+        self.simul = self.qEnd + self.simul
+        self.simul[0] = self.simul[0] + self.m1
+        if self.monthsToForecast > 1:
+            self.simul[1] = self.simul[1] + ((self.simul[0] * self.ARs[0,1]) + (self.finalTwo[1] * self.ARs[0,0]))
+        if self.monthsToForecast > 2:
+            for i in range(2, self.monthsToForecast):
+                self.simul[i] = np.dot(self.ARs, self.simul[(i-2):i])
+        self.finalDistibution = np.exp(self.simul.sum(axis=0) + self.finalSales)
+        self.dfPlot = pd.DataFrame({'Sales': self.finalDistibution, 'Plan': 'Achieved'})
+        self.dfPlot.loc[self.dfPlot.Sales < self.plan, 'Plan'] = 'Not achieved'
+
+        self.left_x = np.quantile(self.finalDistibution, 0.001)
+        self.right_x = np.quantile(self.finalDistibution, 0.99)
+        self.position = (self.plan - self.left_x) / (self.right_x - self.left_x)
+        self.percent_not_achieved = 100*(self.finalDistibution < self.plan).mean()
+        self.density = stats.kde.gaussian_kde(self.dfPlot.Sales)
+        self.x1 = np.linspace(self.left_x, self.plan, 1000)
+        self.x2 = np.linspace(self.plan, self.right_x, 1000)
+
+    def plot(self) -> None:
+        if not hasattr(self, 'model_fit'):
+            raise ValueError("Before calling method 'plot' of an object of class SalesPlanSuccess you first have to fit the ARIMA model by calling method 'fit' and then conduct a simulation by calling method 'simulate' of the same object")
+        if not hasattr(self, 'simul'):
+            raise ValueError("Before calling method 'plot' of an object of class SalesPlanSuccess you first have to conduct a simulation by calling method 'simulate' of the same object")
+        self.ax = plt.gca()
+        plt.fill_between(self.x1, self.density(self.x1), color='orange')
+        plt.fill_between(self.x2, self.density(self.x2), color='green')
+        plt.text(self.position-0.01, 0.2, "Not achieved\n%.1f" % (self.percent_not_achieved,) + '%', color='black', fontsize = 15, transform=self.ax.transAxes, horizontalalignment='right')
+        plt.text(self.position+0.01, 0.7, "Achieved\n%.1f" % (100 - self.percent_not_achieved,) + '%', color='black', fontsize = 15, transform=self.ax.transAxes)
+        plt.xlim(self.left_x, self.right_x)
+        plt.ylim(0)
+        plt.box(False)
+        plt.yticks([])
+        plt.show()

setup.cfg

@@ -0,0 +1,2 @@
+[metadata]
+description-file = README.md

setup.py

@@ -0,0 +1,14 @@
+from distutils.core import setup
+
+setup(
+    name = 'salesplansuccess',
+    packages = ['salesplansuccess'],
+    version = '0.0.1',
+    description = 'This package helps estimate the probability of achieving the annual sales plan for a specific product based on historical monthly sales data with the help of ARIMA modeling and Monte Carlo simulation',
+    author = 'Alexander Yuryatin',
+    author_email = '',
+    url = 'github package source url',
+    download_url = 'download link you saved',
+    keywords = ['sales', 'forecast', 'arima', 'simulation', 'pharma', 'medications', 'pharmacy', 'pharmaceutical', 'ics', 'offtake'],
+    classifiers = [],
+)