Translator Disclaimer
May 1996 A consistent model selection procedure for Markov random fields based on penalized pseudolikelihood
Chuanshu Ji, Lynne Seymour
Ann. Appl. Probab. 6(2): 423-443 (May 1996). DOI: 10.1214/aoap/1034968138

Abstract

Motivated by applications in texture synthesis, we propose a model selection procedure for Markov random fields based on penalized pseudolikelihood. The procedure is shown to be consistent for choosing the true model, even for Gibbs random fields with phase transitions. As a by-product, rates for the restricted mean-square error and moderate deviation probabilities are derived for the maximum pseudolikelihood estimator. Some simulation results are presented for the selection procedure.

Citation

Download Citation

Chuanshu Ji. Lynne Seymour. "A consistent model selection procedure for Markov random fields based on penalized pseudolikelihood." Ann. Appl. Probab. 6 (2) 423 - 443, May 1996. https://doi.org/10.1214/aoap/1034968138

Information

Published: May 1996
First available in Project Euclid: 18 October 2002

zbMATH: 0856.62082
MathSciNet: MR1398052
Digital Object Identifier: 10.1214/aoap/1034968138

Subjects:
Primary: 62M40
Secondary: 62F12, 68U10

Rights: Copyright © 1996 Institute of Mathematical Statistics

JOURNAL ARTICLE
21 PAGES


SHARE
Vol.6 • No. 2 • May 1996
Back to Top