Efficient simulation of Bayesian logistic regression models
In this paper we highlight a data augmentation approach to inference in the Bayesian logistic regression model. We demonstrate that the resulting conditional likelihood of the regression coefficients is multivariate normal, equivalent to a standard Bayesian linear regression, which allows for efficient simulation using a block Gibbs sampler. We illustrate that the method is particularly suited to problems in covariate set uncertainty and random effects models
Year of publication: |
2003
|
---|---|
Authors: | Holmes, C ; Knorr-Held, L |
Publisher: |
München : Ludwig-Maximilians-Universität München, Sonderforschungsbereich 386 - Statistische Analyse diskreter Strukturen |
Subject: | Regression | Theorie | Auxiliary variables | Bayesian logistic regression | Data augmentation | Markov chain Monte Carlo | Model averaging | Random effects |
Saved in:
freely available
Series: | Discussion Paper ; 306 |
---|---|
Type of publication: | Book / Working Paper |
Type of publication (narrower categories): | Working Paper |
Language: | English |
Other identifiers: | 10.5282/ubm/epub.1688 [DOI] 477705332 [GVK] hdl:10419/23865 [Handle] |
Source: |
Persistent link: https://www.econbiz.de/10010263505
Saved in favorites
Similar items by subject
-
State space mixed models for longitudinal obsservations with binary and binomial responses
Czado, Claudia, (2006)
-
State space mixed models for longitudinal observations with binary and binomial responses
Czado, Claudia, (2008)
-
Variable Selection and Functional Form Uncertainty in Cross-Country Growth Regressions
Salimans, Tim, (2011)
- More ...