AUTO-ARIMA: As manual selection of the ARIMA parameters might not lead to discovering the optimal model specification, there is a library called pmdarima (which ports the functionalities of the famous R package called forecast to Python). The key class of the library is called auto_arima and it automatically fits the best model for our time series.
In order to do so, we need to introduce a metric that the function will optimize. A popular choice is the Akaike Information Criterion (AIC), which provides a trade-off between the goodness of fit of the model and its simplicity —AIC deals with the risks of overfitting and underfitting. When we compare multiple models, the lower the value of AIC, the better the model.
auto_arima ...
