Auto-regressive moving averages and other time series models
The model that we tried earlier was a relatively simple pure auto-regressive model. However, we are not stuck with using auto-regression or pure auto-regression alone in our time series models. As with other classes of machine learning models covered in this book, there is a whole zoo of time series techniques, and we cannot cover them all here. However, we did want to mention a few notable techniques that you could explore as you follow up on this material.
Auto-regressive models are often combined with models called moving average models. When these are combined, they are often referred to as auto-regressive moving average (ARMA) or auto-regressive integrated moving average (ARIMA) models. The moving average part of ARMA/ARIMA models allows you to capture the effects of things like white noise or other error terms in your time series, which would actually improve our AR(2) model for air passengers.
Unfortunately, at the time of...
