Autoregressive model orders for Durbin's MA and ARMA estimators

被引：32

作者：

Broersen, PMT ^{[1
]}

机构：

[1] Delft Univ Technol, Dept Appl Phys, Delft, Netherlands

来源：

IEEE TRANSACTIONS ON SIGNAL PROCESSING | 2000年 / 48卷 / 08期

关键词：

D O I：

10.1109/78.852025

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

Durbin's methods 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.

引用

页码：2454 / 2457

页数：4

共 12 条

[1] NEW LOOK AT STATISTICAL-MODEL IDENTIFICATION
AKAIKE, H
[J]. IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 1974, AC19 (06) : 716 - 723
[2] Broersen P. M. T., 1996, Signal Processing VIII, Theories and Applications. Proceedings of EUSIPCO-96, Eighth European Signal Processing Conference, P799
[3] The quality of models for ARMA processes
Broersen, PMT
[J]. IEEE TRANSACTIONS ON SIGNAL PROCESSING, 1998, 46 (06) : 1749 - 1752
[4] BROERSEN PMT, 1992, P 9 IFAC IFORS S ID, P983
[5] Choi B., 1992, ARMA Model Identification
[6] PROBLEMS WITH THE ESTIMATION OF MOVING AVERAGE PROCESSES
DAVIDSON, JEH
[J]. JOURNAL OF ECONOMETRICS, 1981, 16 (03) : 295 - 310
[7] EFFICIENT ESTIMATION OF PARAMETERS IN MOVING-AVERAGE MODELS
DURBIN, J
[J]. BIOMETRIKA, 1959, 46 (3-4) : 306 - 316
[8] Durbin J., 1960, ECONOMETRICA, V28, P233
[9] Findley D. F., 1984, J TIME SER ANAL, V5, P213
[10] ANALYSIS AND SELECTION OF VARIABLES IN LINEAR-REGRESSION
HOCKING, RR
[J]. BIOMETRICS, 1976, 32 (01) : 1 - 49

← 1 2 →