Multi-Step Estimation For Forecasting

Preprint
Publisher Website
Google Scholar
    • preprint
      • Published in RePEc
Abstract
The authors delineate conditions which favor multistep, or dynamic, estimation for multistep forecasting. An analytical example shows how dynamic estimation may accommodate incorrectly specified models as the forecast lead alters, improving forecast performance for some misspecifications. However, in correctly specified models, reducing finite-sample biases does not justify dynamic estimation. In a Monte Carlo forecasting study for integrated processes, estimating a unit root in the presence of a neglected negative moving-average error may favor dynamic estimation, though other solutions exist to that scenario. A second Monte Carlo study obtains the estimator biases and explains those using asymptotic approximations. Copyright 1996 by Blackwell Publishing Ltd (This abstract was borrowed from another version of this item.)
All Related Versions

This publication has 0 references indexed in Scilit: