论文信息 - A principal component approach to dynamic regression models

A principal component approach to dynamic regression models

Abstract In this paper we introduce a dynamic regression model that states how an output is related to an input allowing future values forecasting. The basic tools to set up this model are the orthogonal decomposition of a discrete time stochastic process by means of its principal components analysis, and the linear regression performed on the principal components of input and output processes. The behaviour of this model is empirically studied on real data, showing that low forecast errors are obtained by using this model. A comparison between such a model and a transfer function one, for a particular two time-series case, is discussed.

Mariano J. Valderrama | M. Valderrama | Maria Jose del Moral | M. J. D. Moral

[1] Jean-Claude Deville. Analyse et prévision des séries chronologiques multiples non stationnaires , 1978 .

[2] J. G. Bryan,et al. STATISTICAL METHODS IN FORECASTING , 1962 .

[3] Mariano J. Valderrama,et al. Selected Topics on Stochastic Modelling , 1994 .

[4] Deville. Méthodes statistiques et numériques de l'analyse harmonique , 1974 .

[5] Bovas Abraham,et al. Forecast Functions Implied by Autoregressive Integrated Moving Average Models and Other Related Forecast Procedures , 1986 .

[6] M. A. Wincek. Forecasting With Dynamic Regression Models , 1993 .

[7] J. Ledolter,et al. Parsimony and Its Importance in Time Series Forecasting , 1981 .

[8] Steven C. Hillmer,et al. Likelihood Function of Stationary Multiple Autoregressive Moving Average Models , 1979 .

[9] P. W. Strike,et al. Forecasting and control , 1991 .

[10] P. Young,et al. Time series analysis, forecasting and control , 1972, IEEE Transactions on Automatic Control.