论文信息 - Autoregressive model orders for Durbin's MA and ARMA estimators

Autoregressive model orders for Durbin's MA and ARMA estimators

Durbin's methods (1959, 1960) for moving average (MA) and autoregressive-moving average (ARMA) estimation use the parameters of a long AR model to compute the MA parameters. Linear regression theory is applied to find the best AR order. This yields two different orders: one for the best predicting AR model and another one for the long AR model with the best parameter accuracy, as intermediate for Durbin's estimates. Both orders increase with the sample size and have no finite limiting value.

Piet M. T. Broersen | P. Broersen

[1] James Durbin,et al. The fitting of time series models , 1960 .

[2] ByoungSeon Choi,et al. Arma Model Identification , 1992 .

[3] James Davidson,et al. Problems with the estimation of moving average processes , 1981 .

[4] R. P. Mentz. Estimation in the first-order moving average model through the finite autoregressive approximation : Some asymptotic results , 1977 .

[5] H. Akaike. A new look at the statistical model identification , 1974 .

[6] Piet M.T. Broersen,et al. Small Sample MA Estimation with Durbin’s Method , 1991 .

[7] Piet M. T. Broersen,et al. The best order of long autoregressive models for moving average estimation , 1996, 1996 8th European Signal Processing Conference (EUSIPCO 1996).

[8] D. Findley. ON SOME AMBIGUITIES ASSOCIATED WITH THE FITTING OF ARMA MODELS TO TIME SERIES , 1984 .

[9] Piet M. T. Broersen,et al. The quality of models for ARMA processes , 1998, IEEE Trans. Signal Process..

[10] B. Hofmann-Wellenhof,et al. Introduction to spectral analysis , 1986 .

[11] R. R. Hocking. The analysis and selection of variables in linear regression , 1976 .

[12] J. Durbin. EFFICIENT ESTIMATION OF PARAMETERS IN MOVING-AVERAGE MODELS , 1959 .