Asymptotic mean-squared forecast error when an autoregression with linear trend is fitted to data generated by an I(0) or I(1) process

Abstract.  Assume that a time series is generated by an autoregression which has atmost one unit root. A correctly specified model, including linear time trend, is estimated by ordinary least squares, but no allowance is made for any unit root in the generating process. We investigate the impact of estimation error on the mean-squared error of forecasts calculated from the fitted model.