Assessing prediction error in autoregressive models
HTML articles powered by AMS MathViewer
- by Ping Zhang and Paul Shaman PDF
- Trans. Amer. Math. Soc. 347 (1995), 627-637 Request permission
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
- Hirotugu Akaike, Statistical predictor identification, Ann. Inst. Statist. Math. 22 (1970), 203–217. MR 286233, DOI 10.1007/BF02506337
- H. Akaike, Information theory and an extension of the maximum likelihood principle, Second International Symposium on Information Theory (Tsahkadsor, 1971), Akadémiai Kiadó, Budapest, 1973, pp. 267–281. MR 0483125
- T. W. Anderson and Raúl P. Mentz, On the structure of the likelihood function of autoregressive and moving average models, J. Time Ser. Anal. 1 (1980), no. 2, 83–94. MR 622137, DOI 10.1111/j.1467-9892.1980.tb00302.x
- David R. Brillinger, Some history of the study of higher-order moments and spectra, Statist. Sinica 1 (1991), no. 2, 465–476. MR 1130128
- Wayne A. Fuller and David P. Hasza, Properties of predictors for autoregressive time series, J. Amer. Statist. Assoc. 76 (1981), no. 373, 155–161. MR 608187, DOI 10.1080/01621459.1981.10477622
- Iain Johnstone, On inadmissibility of some unbiased estimates of loss, Statistical decision theory and related topics, IV, Vol. 1 (West Lafayette, Ind., 1986) Springer, New York, 1988, pp. 361–379. MR 927112
- H. Linhart and W. Zucchini, Model selection, Wiley Series in Probability and Mathematical Statistics: Applied Probability and Statistics, John Wiley & Sons, Inc., New York, 1986. MR 866144
- Paul Shaman, Properties of estimates of the mean square error of prediction in autoregressive models, Studies in econometrics, time series, and multivariate statistics, Academic Press, New York, 1983, pp. 331–342. MR 738660
- Paul Shaman and Robert A. Stine, The bias of autoregressive coefficient estimators, J. Amer. Statist. Assoc. 83 (1988), no. 403, 842–848. MR 963814, DOI 10.1080/01621459.1988.10478672
- Ritei Shibata, Selection of the order of an autoregressive model by Akaike’s information criterion, Biometrika 63 (1976), no. 1, 117–126. MR 403130, DOI 10.1093/biomet/63.1.117
- T. P. Speed, What is an analysis of variance?, Ann. Statist. 15 (1987), no. 3, 885–941. With a discussion and a reply by the author. MR 902237, DOI 10.1214/aos/1176350472
Additional Information
- © Copyright 1995 American Mathematical Society
- Journal: Trans. Amer. Math. Soc. 347 (1995), 627-637
- MSC: Primary 62M10; Secondary 62M20
- DOI: https://doi.org/10.1090/S0002-9947-1995-1277143-9
- MathSciNet review: 1277143