Introduction to R for Quantitative Finance

Linear time series modeling and forecasting

An important class of linear time series models is the family of Autoregressive Integrated Moving Average (ARIMA) models, proposed by Box and Jenkins (1976). It assumes that the current value can depend only on the past values of the time series itself or on past values of some error term.

According to Box and Jenkins, building an ARIMA model consists of three stages:

1. Model identification.
2. Model estimation.
3. Model diagnostic checking.

The model identification step involves determining the order (number of past values and number of past error terms to incorporate) of a tentative model using either graphical methods or information criteria. After determining the order of the model, the parameters of the model ...