Discretizing a Normal Prior for Change Point Estimation in Switching Regressions

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Bayes decision procedures are considered for change point estimation in the simple bilinear segmented model. A discretized normal prior density is employed as the prior distribution for the change point index. Posterior probability functions are developed for this index under a vague prior formulation on the regression parameters. The procedure is applied to an example involving mercury toxicity data.

Original languageEnglish (US)
Pages (from-to)777-782
Number of pages6
JournalBiometrical Journal
Volume29
Issue number7
DOIs
StatePublished - 1987
Externally publishedYes

Fingerprint

Change-point Estimation
Regression
Bayes Procedures
Mercury
Probability function
Change Point
Posterior Probability
Decision Procedures
Toxicity
Prior distribution
Formulation
Switching regression
Change point
Model

Keywords

  • Bayes analysis
  • Nonlinear regression
  • Segmented regression

ASJC Scopus subject areas

  • Statistics and Probability
  • Medicine(all)
  • Statistics, Probability and Uncertainty

Cite this

Discretizing a Normal Prior for Change Point Estimation in Switching Regressions. / Piegorsch, Walter W.

In: Biometrical Journal, Vol. 29, No. 7, 1987, p. 777-782.

Research output: Contribution to journalArticle

@article{3a2aeed0a88c452e837c3056c4e4fc58,
title = "Discretizing a Normal Prior for Change Point Estimation in Switching Regressions",
abstract = "Bayes decision procedures are considered for change point estimation in the simple bilinear segmented model. A discretized normal prior density is employed as the prior distribution for the change point index. Posterior probability functions are developed for this index under a vague prior formulation on the regression parameters. The procedure is applied to an example involving mercury toxicity data.",
keywords = "Bayes analysis, Nonlinear regression, Segmented regression",
author = "Piegorsch, {Walter W}",
year = "1987",
doi = "10.1002/bimj.4710290705",
language = "English (US)",
volume = "29",
pages = "777--782",
journal = "Biometrical Journal",
issn = "0323-3847",
publisher = "Wiley-VCH Verlag",
number = "7",

}

TY - JOUR

T1 - Discretizing a Normal Prior for Change Point Estimation in Switching Regressions

AU - Piegorsch, Walter W

PY - 1987

Y1 - 1987

N2 - Bayes decision procedures are considered for change point estimation in the simple bilinear segmented model. A discretized normal prior density is employed as the prior distribution for the change point index. Posterior probability functions are developed for this index under a vague prior formulation on the regression parameters. The procedure is applied to an example involving mercury toxicity data.

AB - Bayes decision procedures are considered for change point estimation in the simple bilinear segmented model. A discretized normal prior density is employed as the prior distribution for the change point index. Posterior probability functions are developed for this index under a vague prior formulation on the regression parameters. The procedure is applied to an example involving mercury toxicity data.

KW - Bayes analysis

KW - Nonlinear regression

KW - Segmented regression

UR - http://www.scopus.com/inward/record.url?scp=33645080837&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33645080837&partnerID=8YFLogxK

U2 - 10.1002/bimj.4710290705

DO - 10.1002/bimj.4710290705

M3 - Article

VL - 29

SP - 777

EP - 782

JO - Biometrical Journal

JF - Biometrical Journal

SN - 0323-3847

IS - 7

ER -