Direct and iterated multistep AR methods for difference stationary processes

Direct and iterated multistep AR methods for difference stationary processes

0.00 Avg rating0 Votes
Article ID: iaor20112084
Volume: 27
Issue: 2
Start Page Number: 266
End Page Number: 280
Publication Date: Apr 2011
Journal: International Journal of Forecasting
Authors:
Keywords: stochastic processes
Abstract:

The paper focuses on a comparison between the direct and iterated AR predictors for difference stationary processes. In particular, it provides new methods for comparing the efficiency of the two predictors. The methods are based on an encompassing representation for the two predictors, which enables us to derive their properties quite easily under a maintained model. The paper provides an analytical expression for the mean square forecast error of the two predictors and derives useful recursive formulae for computing the direct and iterated coefficients. From an empirical standpoint, we propose estimators of the AR coefficients based on the tapered Yule‐Walker estimates; we also provide a test of equal forecast accuracy which is very simple to implement and whose critical values are obtained using the bootstrap method.

Reviews

Required fields are marked *. Your email address will not be published.