A Symmetric prior for multinomial probit models

DOI10.1214/20-BA1233arXiv1912.10334OpenAlexW3072725072MaRDI QIDQ6201436FDOQ6201436

Authors: Lane F. Burgette, David W. Puelz, P. Richard Hahn 

Publication date: 20 February 2024

Published in: Bayesian Analysis (Search for Journal in Brave)

Abstract: Fitted probabilities from widely used Bayesian multinomial probit models can depend strongly on the choice of a base category, which is used to uniquely identify the parameters of the model. This paper proposes a novel identification strategy, and associated prior distribution for the model parameters, that renders the prior symmetric with respect to relabeling the outcome categories. The new prior permits an efficient Gibbs algorithm that samples rank-deficient covariance matrices without resorting to Metropolis-Hastings updates.

Full work available at URL: https://arxiv.org/abs/1912.10334

zbMATH Keywords

Gibbs samplerdiscrete choicebase categorysum-to-zero identification

Mathematics Subject Classification ID

Statistics (62-XX)

Cites Work

• A Bayesian analysis of the multinomial probit model using marginal data augmentation
• A Bayesian analysis of the multinomial probit model with fully identified parameters
• The Calculation of Posterior Distributions by Data Augmentation
• Bayesian Analysis of Binary and Polychotomous Response Data
• Bayesian Statistics and Marketing
• Discrete Choice Methods with Simulation
• A Bayesian multinomial probit model for the analysis of panel choice data
• Parsimonious estimation of the covariance matrix in multinomial probit models

Cited In (2)

• Globally and symmetrically identified Bayesian multinomial probit model
• Fast Variational Bayes Methods for Multinomial Probit Models

This page was built for publication: A Symmetric prior for multinomial probit models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6201436)