Bayesian model selection in ordinal quantile regression

详细信息    查看全文

• 作者：Rahim Alhamzawi ^{ralhamzawi@yahoo.com" class="auth_mail" title="E-mail the corresponding author}
• 关键词：Bayesian inference ; MCMC ; Quantile regression ; Ordinal models ; SSVS
• 刊名：Computational Statistics & Data Analysis
• 出版年：2016
• 出版时间：November 2016
• 年：2016
• 卷：103
• 期：Complete
• 页码：68-78
• 全文大小：600 K

文摘

A Bayesian stochastic search variable selection (BSSVS) method is presented for variable selection in quantile regression (QReg) for ordinal models. A Markov Chain Monte Carlo (MCMC) method is adopted to draw the unknown quantities from the full posteriors. Through simulations and analysis of an educational attainment dataset, the performance of the proposed approach is compared with some existing approaches, showing that the proposed approach performs quite good in comparison to some other methods.