An autoregressive (AR) model is a time series model that regresses the dependent variable against one or more lagged values of itself. We use the past values of a variable to predict the current and future values of the variable.
If the AR model only regresses on itself with a lag of one period, we call this a first order autoregressive model AR(1).
If it has a lag of one and two periods, we call this a second-order autoregressive model AR(2).
In general, an AR model of order p is expressed as such, where p indicates the number of lagged values that the autoregressive model will include as independent variables.
See also: Mean reversion, Covariance stationarity« Back to Index