main.py

# -*- coding: utf-8 -*-
"""
Created on Wed Nov  4 23:51:34 2020

@author: Yu Zhe
"""
from forecast_arima import Predictor_ARIMA
from forecast_prophet import Predictor_fbprophet
from forecast_arma import Predictor_ARMA
from forecast_hwes import Predictor_HWES

import pandas as pd
df = pd.read_csv("../monthly-car-sales.csv")

#%%
"""
ARIMA
"""
arima_result = Predictor_ARIMA(df)
arima_result.fit_model()
arima_result.evaluate_model(output=False)
mse_arima,rmse_arima,mae_arima=arima_result.analyze_estimator(output=False)
print(mse_arima,rmse_arima,mae_arima)
arima_pred = arima_result.outsample_forecast(output=False)
print(arima_pred)
#arima_pred.plot()
arima_result.diagnostic_model()

#%%
"""
Fbprophet
"""
fbprophet_result = Predictor_fbprophet(df)
fbprophet_pred = fbprophet_result.outsample_forecast(output=False)
print(fbprophet_pred)
print(fbprophet_result.evaluate_model(output=False))
fbprophet_pred_yhat = fbprophet_pred['yhat']
fbprophet_pred_yhat.index = fbprophet_pred['ds']

#%%
"""
Get Seasonality
"""
fbprophet_seasonality = fbprophet_pred['additive_terms']
fbprophet_seasonality.index = fbprophet_pred['ds']
fbprophet_seasonality.plot()

#%%
"""
ARMA
"""
arma_result = Predictor_ARMA(df)
arma_result.evaluate_model(output=False)
print(arma_result)
arma_pred = arma_result.outsample_forecast(output=False)
#arma_pred.plot()
arma_pred+=fbprophet_seasonality

#%%
"""
HWES
"""
hwes_result = Predictor_HWES(df)
hwes_result.param_selection(12) #Choose best config
mse_hwes,rmse_hwes,mae_hwes=hwes_result.evaluate_model(output=False)
print(mse_hwes,rmse_hwes,mae_hwes)
hwes_pred = hwes_result.outsample_forecast(output=False)
print(hwes_pred)
#arma_pred.plot()

#%%
"""
Comparison
"""

original_df = df.set_index('ds')
ax = original_df['y'].plot(label= 'Observed', color='cyan', legend=True)

arima_pred.plot(label= 'ARIMA_forecast', color='blue', legend=True, ax=ax)
fbprophet_pred_yhat.plot(label= 'fbprophet_forecast', color='red', legend=True, ax=ax)
arma_pred.plot(label= 'ARMA_forecast', color='green', legend=True, ax=ax)
hwes_pred.plot(label= 'HWES_forecast', color='orange', legend=True, ax=ax)

#%%
"""
Evaluation metric for learning model
1. Mean absolute error (MAE)
2. Mean squared error (MSE)
3. Root Mean Square Error (RMSE)
4. Mean absolute percentage error (MAPE)
5. Symmetric mean absolute percentage error (SMAPE)
6. Mean Forecast Error (MFE)
7. Normalized mean squared error (NMSE)
8. Theil's U statistic
"""
from forecast_recommender import forecast_recommender

df1 = pd.read_csv("../monthly-car-sales.csv")
recommender = forecast_recommender(df1)
result, model_name = recommender.auto_forecast(0)
print(model_name)
print(result)