# Start writing code here... import numpy as np import plotly.figure_factory as ff import pandas as pd import datetime

Lunes_1= hist_data = np.random.lognormal(mean=5.5, sigma= 0.18, size= 156) Lunes_2=hist_data = np.random.lognormal(mean=5, sigma= 0.18, size= 156) Lunes_3=hist_data = np.random.lognormal(mean=5, sigma= 0.18, size= 156) Martes_1=hist_data = np.random.lognormal(mean=5, sigma= 0.18, size= 156) Martes_2=hist_data = np.random.lognormal(mean=5, sigma= 0.18, size= 156) Martes_3=hist_data = np.random.lognormal(mean=5, sigma= 0.18, size= 156) Miercoles_1=hist_data = np.random.lognormal(mean=5, sigma= 0.18, size= 156) Miercoles_2=hist_data = np.random.lognormal(mean=4.5, sigma= 0.18, size= 156) Miercoles_3=hist_data = np.random.lognormal(mean=5, sigma= 0.18, size= 156) Jueves_1=hist_data = np.random.lognormal(mean=5, sigma= 0.18, size= 156) Jueves_2=hist_data = np.random.lognormal(mean=4.5, sigma= 0.18, size= 156) Jueves_3=hist_data = np.random.lognormal(mean=5, sigma= 0.18, size= 156) Viernes_1=hist_data = np.random.lognormal(mean=5, sigma= 0.18, size= 156) Viernes_2=hist_data = np.random.lognormal(mean=5, sigma= 0.18, size= 156) Viernes_3=hist_data = np.random.lognormal(mean=5, sigma= 0.18, size= 156) Sabado_1=hist_data = np.random.lognormal(mean=5, sigma= 0.18, size= 156) Sabado_2=hist_data = np.random.lognormal(mean=5, sigma= 0.18, size= 156) Sabado_3=hist_data = np.random.lognormal(mean=5, sigma= 0.18, size= 156) Domingo_1=hist_data = np.random.lognormal(mean=5, sigma= 0.18, size= 156) Domingo_2=hist_data = np.random.lognormal(mean=5.5, sigma= 0.18, size= 156) Domingo_3=hist_data = np.random.lognormal(mean=5, sigma= 0.18, size= 156)

days= ['Lunes', 'Martes', 'Miercoles', 'Jueves', 'Viernes', 'Sábado', 'Domingo']*156 days = pd.DataFrame(pd.Series(days))

morning = np.vstack((Lunes_1, np.vstack((Martes_1, np.vstack((Miercoles_1, np.vstack((Jueves_1, np.vstack((Viernes_1, np.vstack((Sabado_1, Domingo_1 )) )) )) )) )) )).T.reshape((1092,1)) evening = np.vstack((Lunes_2, np.vstack((Martes_2, np.vstack((Miercoles_2, np.vstack((Jueves_2, np.vstack((Viernes_2, np.vstack((Sabado_2, Domingo_2 )) )) )) )) )) )).T.reshape((1092, 1)) night = np.vstack((Lunes_3, np.vstack((Martes_3, np.vstack((Miercoles_3, np.vstack((Jueves_3, np.vstack((Viernes_3, np.vstack((Sabado_3, Domingo_3 )) )) )) )) )) )).T.reshape((1092, 1))

morning.shape

morning_final = pd.concat((pd.DataFrame(morning), days), axis=1) evening_final = pd.concat((pd.DataFrame(evening), days), axis=1) night_final = pd.concat((pd.DataFrame(night), days), axis=1)

date=datetime.date.today() time_delta = datetime.timedelta(hours=35064) First_Date=date-time_delta date_time = [First_Date] aux_time = First_Date for i in range(1091): aux_time = aux_time + datetime.timedelta(hours=24) date_time.append(aux_time)

hist_data = np.random.lognormal(mean=5, sigma= 0.18, size= 50000) #hist_data = [x for x in hist_data if x <= 200] group_labels = ['Martes en la noche', 'Lunes en la Mañana', 'Miercoles por la tarde'] # name of the dataset fig = ff.create_distplot([Martes_3, Lunes_1, Miercoles_2], group_labels) fig.update_layout(title_text='Demandas para una semana') fig.show()

# Using plotly.express import plotly.express as px df = pd.concat((pd.DataFrame(morning), pd.DataFrame(evening), pd.DataFrame(night), pd.DataFrame(days), pd.DataFrame(date_time)), axis=1) df.columns=['Morning', 'Evening', 'Night', 'Weekday', 'Date'] df_aux = df.head(150) fig = px.line(df_aux, x='Date', y=['Morning', 'Evening', 'Night']) fig.show()

!pip install statsmodels

from statsmodels.tsa.arima.model import ARIMA from sklearn.metrics import mean_squared_error from math import sqrt from matplotlib import pyplot # split into train and test sets X = df_morning = df[['Date', 'Morning']] X['Morning'] = pd.to_numeric(df["Morning"], downcast="float") X['Date']= pd.to_datetime(X['Date']) print(X) X.set_index('Date', inplace=True) print(X) X=X.values print(X.shape) size = int(len(X) * 0.66) train, test = X[0:size], X[size:len(X)] history = [x for x in train] predictions = list()

# walk-forward validation for t in range(len(test)): model = ARIMA(history, order=(7,1,7)) model_fit = model.fit() output = model_fit.forecast() yhat = output[0] predictions.append(yhat) obs = test[t] history.append(obs) print('predicted=%f, expected=%f' % (yhat, obs)) print(len(test)) print(t) # evaluate forecasts rmse = sqrt(mean_squared_error(test, predictions)) print('Test RMSE: %.3f' % rmse) # plot forecasts against actual outcomes pyplot.plot(test) pyplot.plot(predictions, color='red') pyplot.show()

rmse = sqrt(mean_squared_error(test, predictions)) print('Test RMSE: %.3f' % rmse) # plot forecasts against actual outcomes pyplot.plot(test) pyplot.plot(predictions, color='red') pyplot.show()

from sklearn.metrics import mean_squared_error from math import sqrt rmse = sqrt(mean_squared_error(test, predictions)) print('Test RMSE: %.3f' % rmse) # plot forecasts against actual outcomes pyplot.plot(test) pyplot.plot(predictions, color='red') pyplot.show()

import statsmodels.api as sm import itertools p = d = q = range(0, 2) pdq = list(itertools.product(p, d, q)) seasonal_pdq = [(x[0], x[1], x[2], 12) for x in list(itertools.product(p, d, q))] print('Examples of parameter combinations for Seasonal ARIMA...') print('SARIMAX: {} x {}'.format(pdq[1], seasonal_pdq[1])) print('SARIMAX: {} x {}'.format(pdq[1], seasonal_pdq[2])) print('SARIMAX: {} x {}'.format(pdq[2], seasonal_pdq[3])) print('SARIMAX: {} x {}'.format(pdq[2], seasonal_pdq[4]))