Information and Posterior Probability Criteria for Model Selection in Local Likelihood Estimation

Rafael A. Irizarry

Research output: Contribution to journalArticle

Abstract

Local likelihood estimation has proven to be an effective method for obtaining estimates of parameters that vary with a covariate. To obtain useful estimates of such parameters, approximating models are used. In such cases it is useful to consider window based estimates. We may need to choose between competing approximating models. In this article, we propose a modification to the methods used to motivate many information and posterior probability criteria for the weighted likelihood case. We derive weighted versions for two of the most widely known criteria, namely the AIC and BIC. Via a simple modification, the criteria are also made useful for window span selection. The usefulness of the weighted version of these criteria is demonstrated through a simulation study and an application to three datasets.

Original languageEnglish (US)
Pages (from-to)303-315
Number of pages13
JournalJournal of the American Statistical Association
Volume96
Issue number453
StatePublished - Mar 2001

Fingerprint

Local Likelihood
Posterior Probability
Model Selection
Estimate
Weighted Likelihood
Covariates
Choose
Simulation Study
Vary
Likelihood estimation
Posterior probability
Model selection
Model

Keywords

  • Information criteria
  • Local likelihood
  • Local regression
  • Model selection
  • Posterior probability criteria
  • Signal processing
  • Weighted Kullback-Leibler
  • Window size selection

ASJC Scopus subject areas

  • Mathematics(all)
  • Statistics and Probability

Cite this

Information and Posterior Probability Criteria for Model Selection in Local Likelihood Estimation. / Irizarry, Rafael A.

In: Journal of the American Statistical Association, Vol. 96, No. 453, 03.2001, p. 303-315.

Research output: Contribution to journalArticle

@article{000b7c54154841bda257f5246e1e8946,
title = "Information and Posterior Probability Criteria for Model Selection in Local Likelihood Estimation",
abstract = "Local likelihood estimation has proven to be an effective method for obtaining estimates of parameters that vary with a covariate. To obtain useful estimates of such parameters, approximating models are used. In such cases it is useful to consider window based estimates. We may need to choose between competing approximating models. In this article, we propose a modification to the methods used to motivate many information and posterior probability criteria for the weighted likelihood case. We derive weighted versions for two of the most widely known criteria, namely the AIC and BIC. Via a simple modification, the criteria are also made useful for window span selection. The usefulness of the weighted version of these criteria is demonstrated through a simulation study and an application to three datasets.",
keywords = "Information criteria, Local likelihood, Local regression, Model selection, Posterior probability criteria, Signal processing, Weighted Kullback-Leibler, Window size selection",
author = "Irizarry, {Rafael A.}",
year = "2001",
month = "3",
language = "English (US)",
volume = "96",
pages = "303--315",
journal = "Journal of the American Statistical Association",
issn = "0162-1459",
publisher = "Taylor and Francis Ltd.",
number = "453",

}

TY - JOUR

T1 - Information and Posterior Probability Criteria for Model Selection in Local Likelihood Estimation

AU - Irizarry, Rafael A.

PY - 2001/3

Y1 - 2001/3

N2 - Local likelihood estimation has proven to be an effective method for obtaining estimates of parameters that vary with a covariate. To obtain useful estimates of such parameters, approximating models are used. In such cases it is useful to consider window based estimates. We may need to choose between competing approximating models. In this article, we propose a modification to the methods used to motivate many information and posterior probability criteria for the weighted likelihood case. We derive weighted versions for two of the most widely known criteria, namely the AIC and BIC. Via a simple modification, the criteria are also made useful for window span selection. The usefulness of the weighted version of these criteria is demonstrated through a simulation study and an application to three datasets.

AB - Local likelihood estimation has proven to be an effective method for obtaining estimates of parameters that vary with a covariate. To obtain useful estimates of such parameters, approximating models are used. In such cases it is useful to consider window based estimates. We may need to choose between competing approximating models. In this article, we propose a modification to the methods used to motivate many information and posterior probability criteria for the weighted likelihood case. We derive weighted versions for two of the most widely known criteria, namely the AIC and BIC. Via a simple modification, the criteria are also made useful for window span selection. The usefulness of the weighted version of these criteria is demonstrated through a simulation study and an application to three datasets.

KW - Information criteria

KW - Local likelihood

KW - Local regression

KW - Model selection

KW - Posterior probability criteria

KW - Signal processing

KW - Weighted Kullback-Leibler

KW - Window size selection

UR - http://www.scopus.com/inward/record.url?scp=1842765676&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=1842765676&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:1842765676

VL - 96

SP - 303

EP - 315

JO - Journal of the American Statistical Association

JF - Journal of the American Statistical Association

SN - 0162-1459

IS - 453

ER -