Small-Sample Properties of Predictions from the Regression Model with Autoregressive Errors

Abstract
Monte Carlo methods are used to examine the small-sample properties of various estimators of asymptotic prediction variance (APV) from the regression model with autoregressive errors. Two practical estimators of APV, one of which includes terms reflecting parameter estimation and one of which excludes these terms, are compared to the mean squared error of prediction for regression models with different autocorrelation of the errors, different sample sizes, and different period-ahead predictions. The inclusion of terms reflecting the estimation of parameters is found to be worthwhile, particularly in small samples.

This publication has 0 references indexed in Scilit: