Bias correction of OLSE in the regression model with lagged dependent variables. (Q1583509)

It is well known that the ordinary least-squares estimates (OLSE) of autoregressive models are biased in small samples. In this paper, an attempt is made to obtain unbiased estimates in the sense of median or mean. Using Monte Carlo simulation techniques, we extend the median-unbiased estimator proposed by \textit{D. W. K. Andrews} [Econometrica 61, 139--165 (1993; Zbl 0772.62064)] to higher-order autoregressive processes, nonnormal error terms and inclusion of any exogenous variables. Also, we introduce the mean-unbiased estimator, which is compared with OLSE and a medium-unbiased estimator. Some simulation studies are performed to examine whether the proposed estimation procedure works well or not, where AR(\(p)\) for \(p=1,2,3\) models are examined. We obtain the result that it is possible to recover the true parameter values from OLSE and that the proposed procedure gives less-biased estimators than OLSE. Finally, using actually obtained data, an empirical example of the median- and mean-unbiased estimators are shown.