Bias of some commonly-used time series estimates
- 1 August 1983
- journal article
- research article
- Published by Oxford University Press (OUP) in Biometrika
- Vol. 70 (2) , 389-399
- https://doi.org/10.1093/biomet/70.2.389
Abstract
We study the bias of Yule-Walker and least squares estimates for univariate and multivariate autoregressive processes. We obtain explicit formulae for the large-sample bias of YuleWalker estimates in the scalar first- and second-order cases and for least squares estimates in the general case. Both simulations and theory indicate that Yule-Walker estimates are inferior to least squares estimates. For strongly autocorrelated processes, Yule-Walker estimates can be severely biased even for comparatively large-sample sizes.Keywords
This publication has 0 references indexed in Scilit: