Asymptotically efficient autoregressive model selection for multistep prediction

A direct method for multistep prediction of a stationary time series involves fitting, by linear regression, a different autoregression for each lead time, h, and to select the order to be fitted, % MathType!MTEF!2!1!+-% feaafeart1ev1aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn% hiov2DGi1BTfMBaeXafv3ySLgzGmvETj2BSbqefm0B1jxALjhiov2D% aebbfv3ySLgzGueE0jxyaibaiGc9yrFr0xXdbba91rFfpec8Eeeu0x% Xdbba9frFj0-OqFfea0dXdd9vqaq-JfrVkFHe9pgea0dXdar-Jb9hs% 0dXdbPYxe9vr0-vr0-vqpWqaaeaabiGaaiaacaqabeaadaqaaqGaaO% qaaiqadUgagaacamaaBaaaleaacaWGObaabeaaaaa!3E44!\[\tilde k_h\], from the data. By contrast, a more usual ‘plug-in’ method involves the least-squares fitting of an initial k-th order autoregression, with k itself selected by an order selection criterion. A bound for the mean squared error of prediction of the direct method is derived and employed for defining an asymptotically efficient order selection for h-step prediction, h > 1; the Sh(k) criterion of Shibata (1980) is asymptotically efficient according to this definition. A bound for the mean squared error of prediction of the plug-in method is also derived and used for a comparison of these two alternative methods of multistep prediction. Examples illustrating the results are given.

[1]  H. Akaike.  Statistical predictor identification , 1970 .

[2]  G. Reinsel,et al.  Prediction of multivariate time series by autoregressive model fitting , 1985 .

[3]  C. Granger,et al.  Forecasting from non-linear models in practice , 1994 .

[4]  D. Cox.  Prediction by Exponentially Weighted Moving Averages and Related Methods , 1961 .

[5]  R. Shibata.  Asymptotically Efficient Selection of the Order of the Model for Estimating Parameters of a Linear Process , 1980 .

[6]  Taku Yamamoto,et al.  Asymptotic mean square prediction error for an autoregressive model with estimated coefficients , 1976 .

[7]  R. Bhansali,et al.  Convergence of Moments of Least Squares Estimators for the Coefficients of an Autoregressive Process of Unknown Order , 1991 .

[8]  Bonnie K. Ray,et al.  MODELING LONG‐MEMORY PROCESSES FOR OPTIMAL LONG‐RANGE PREDICTION , 1993 .

[9]  S. Unnikrishna Pillai,et al.  A new spectrum extension method that maximizes the multistep minimum prediction error-generalization of the maximum entropy concept , 1992, IEEE Trans. Signal Process..

[10]  H. Akaike.  Statistical predictor identification , 1970 .

[11]  R. Shibata.  An Optimal Autoregressive Spectral Estimate , 1981 .

[12]  G. C. Tiao,et al.  Robustness of maximum likelihood estimates for multi-step predictions: The exponential smoothing case , 1993 .

[13]  David R. Brillinger,et al.  Time Series: Data Analysis and Theory. , 1982 .

[14]  R. Tempo,et al.  Optimal algorithms theory for robust estimation and prediction , 1985 .

[15]  R. Bhansali.  Effects of Not Knowing the Order of an Autoregressive Process on the Mean Squared Error of Prediction—I , 1981 .

[16]  Petre Stoica,et al.  Uniqueness of estimated k-step prediction models of ARMA processes , 1984 .

[17]  P. Whittle.  Prediction and Regulation by Linear Least-Square Methods , 1983 .

[18]  Edoardo Mosca,et al.  Performance improvements of self-tuning controllers by multistep horizons: The MUSMAR approach , 1984, Autom..

[19]  U. Grenander,et al.  An extension of a theorem of G. Szegö and its application to the study of stochastic processes , 1954 .

[20]  M. B. Priestley,et al.  The Development and Construction of General Nonlinear Models in Time Series Analysis , 1985 .

[21]  Paul Kabaila,et al.  Estimation based on one step ahead prediction versus estimation based on multi-step ahead prediction , 1981 .

[22]  Genshiro Kitagawa,et al.  An approach to the prediction of time series with trends and seasonalities , 1982, 1982 21st IEEE Conference on Decision and Control.

[23]  R. J. Bhansali,et al.  Asymptotically Efficient Selection of the Order by the Criterion Autoregressive Transfer Function , 1986 .

[24]  Andrew A. Weiss,et al.  Multi-step estimation and forecasting in dynamic models , 1991 .

[25]  C. Hurvich.  Automatic selection of a linear predictor through frequency domain cross-validation , 1987 .

[26]  H. Akaike,et al.  Information Theory and an Extension of the Maximum Likelihood Principle , 1973 .

[27]  Gwilym M. Jenkins,et al.  Time series analysis, forecasting and control , 1972 .

[28]  T. Anderson.  Statistical analysis of time series , 1974 .

[29]  G. Baxter.  A norm inequality for a “finite-section” Wiener-Hopf equation , 1963 .

[30]  R. Bhansali.  ESTIMATION OF THE PREDICTION ERROR VARIANCE AND AN R2 MEASURE BY AUTOREGRESSIVE MODEL FITTING , 1993 .

[31]  Arye Nehorai,et al.  On multistep prediction error methods for time series models , 1989 .

[32]  K. Wall,et al.  A preference-based method for forecast combination , 1989 .

[33]  D. B. Preston.  Spectral Analysis and Time Series , 1983 .

[34]  R. Bhansali,et al.  Some properties of the order of an autoregressive model selected by a generalization of Akaike∘s EPF criterion , 1977 .