Multi-step forecasting for long-memory processes

In this paper we present results of a simulation study to assess and compar e the accuracy of forecasting techniques for long-memory processes in small sample sizes. We analyse differences between adaptive ARMA(1,1) L-step for ecasts, where the parameters are estimated by minimizing the sum of squares of L-step forecast errors, and forecasts obtained by using long-memory mod els. We compare widths of the forecast intervals for both methods, and disc uss some computational issues associated with the ARMA(1,1) method. Our res ults illustrate the importance and usefulness of long-memory models for mul ti-step forecasting. Copyright (C) 1999 John Wiley & Sons, Ltd.