boot.residuals.arima {PRTest} | R Documentation |
This function takes a fitted model, generates a simulated time series from that
model and then fits the model using arima
and returns the residuals.
boot.residuals.arima(obj, StableQ = FALSE)
obj |
object of class Arima, that is the output from Arima |
StableQ |
True, use stable innovations. Otherwise Gaussian innovations. |
The fitted model is simulated using SimulateARMA. When StableQ is true, the parameters of the stable distribution are estimated in the original fitted model using the residuals. These parameters are then used to generate innovations for the ARMA model. When StableQ is false, Gaussian innovations are used.
a time series or vector of length n is produced, where n is the length of the residuals in the original fitted model.
A.I. McLeod
J.-W. Lin and McLeod (2007, accepted). Portmanteau Tests for ARMA Models with Infinite Variance. Journal of Time Series Analysis.
data(CRSP) CRSP.AR5<-arima(CRSP, c(5,0,0)) acf(boot.residuals.arima(CRSP.AR5, StableQ=TRUE)