The difference between forecast and predict in python time series model

For example, in the monthly data with a period of 12 from 2017.01.01 to 017.12.31, the model model is obtained by using ARIMA fitting.

model.get_prediction(start='2017.09.01') will get the predicted value of 2017.09.01-2017.12.31 in the sample calculated by the fitting model;

model. get_forcast(step=5) will get the predicted value of the sample extrapolation for 5 periods, that is, the five months from 2018.01.01 to 2018.05.31;

Note:

model.get_prediction can also be done For prediction of extrapolated values, just set the specific termination period.