Publications Meetings The Profession Membership Programs Math Samplings Policy & Advocacy In the News About the AMS
   
Mobile Device Pairing
Green Open Access
Transactions of the American Mathematical Society
Transactions of the American Mathematical Society
ISSN 1088-6850(online) ISSN 0002-9947(print)

Assessing prediction error in autoregressive models


Authors: Ping Zhang and Paul Shaman
Journal: Trans. Amer. Math. Soc. 347 (1995), 627-637
MSC: Primary 62M10; Secondary 62M20
MathSciNet review: 1277143
Full-text PDF Free Access

Abstract | References | Similar Articles | Additional Information

Abstract: Assessing prediction error is a problem which arises in time series analysis. The distinction between the conditional prediction error $ e$ and the unconditional prediction error $ E(e)$ has not received much attention in the literature. Although one can argue that the conditional version is more practical, we show in this article that assessing $ e$ is nearly impossible. In particular, we use the correlation coefficient $ \operatorname{corr} (\hat e,e)$, where $ \hat e$ is an estimate of $ e$, as a measure of performance and show that $ {\lim _{T \to \infty }}\sqrt T \operatorname{corr} (\hat e,e) = C$ where $ T$ is the sample size and $ C > 0$ is some constant. Furthermore, the value of $ C$ is large only when the process is extremely non-Gaussian or nearly nonstationary.


References [Enhancements On Off] (What's this?)


Similar Articles

Retrieve articles in Transactions of the American Mathematical Society with MSC: 62M10, 62M20

Retrieve articles in all journals with MSC: 62M10, 62M20


Additional Information

DOI: http://dx.doi.org/10.1090/S0002-9947-1995-1277143-9
PII: S 0002-9947(1995)1277143-9
Keywords: Conditional prediction error, correlation, cumulant, higher-order spectrum, non-Gaussian model
Article copyright: © Copyright 1995 American Mathematical Society