Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
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