I studied the SARIMAX model: https://www.digitalocean.com/community/tutorials/a-guide-to-time-series-forecasting-with-arima-in-python-3 Data Date Range 1958-2001
mod = sm.tsa.statespace.SARIMAX(y,
order=(1, 1, 1),
seasonal_order=(1, 1, 1, 12),
enforce_stationarity=False,
enforce_invertibility=False)
results = mod.fit()
When I approached the ARIMA time series model, I found that all the data in the date range matches the model parameters. But when checking the forecasts, the author used the date starting from 1998-01-01 as part of the data range for fitting the model.
pred = results.get_prediction(start=pd.to_datetime('1998-01-01'), dynamic=False)
I know that in the machine learning model, the training data and the verification (testing) data are different, I mean a different range. I mean the author is right? why so (I mean the reason to use all the data about the trains), I am new to the SARIMAX model.
, , , , , , , = (1,1,1), season_order = (1, 1, 1, 12). !