On the penalty factor for autoregressive order selection in finite samples

Author

Broersen, P.M.T. ; Wensink, H.E.

Author_Institution

Dept. of Appl. Phys., Delft Univ. of Technol., Netherlands

Volume

44

Issue

3

fYear

1996

fDate

3/1/1996 12:00:00 AM

Firstpage

748

Lastpage

752

Abstract

The order selection criterion that selects models with the smallest squared error of prediction is the best. The finite sample theory describes equivalents for asymptotic order selection criteria that are better in the finite sample practice. This correction for finite sample statistics is the most important. Afterwards, a preference in order selection criteria can be obtained by computing an optimal value for the penalty factor based on a subjective balance of the risks of overfitting and underfitting

Keywords

autoregressive processes; error analysis; parameter estimation; prediction theory; signal sampling; statistical analysis; AR parameter estimation; asymptotic order selection; autoregressive order selection; finite sample statistics; finite sample theory; finite samples; order selection criteria; overfitting; penalty factor; prediction; squared error; underfitting; Adaptive algorithm; Adaptive signal processing; Crosstalk; Equations; Higher order statistics; Lakes; Polynomials; Signal processing; Signal processing algorithms; USA Councils;

fLanguage

English

Journal_Title

Signal Processing, IEEE Transactions on

Publisher

ieee

ISSN

1053-587X

Type

jour

DOI

10.1109/78.489055

Filename

489055