文章基本信息

标题：Linear regression with a dependent skewed Dirichlet process
本地全文：下载
作者：Fernando Quintana
期刊名称：Chilean Journal of Statistics
印刷版ISSN：0718-7912
电子版ISSN：0718-7920
出版年度：2010
卷号：1
期号：2
页码：35-49
出版社：Chilean Statistical Society
摘要：In the context of linear regression models, two strategies for a exible distribution of responses are explored. The two approaches are particular instances of a skewed Dirichlet process with dependence on covariates, so that not only the mean response function but the entire distribution of responses depends on covariates. The two constructions are discussed and compared in the context of two examples
关键词：Bayes factor Nonparametric Bayes Predictive inference Skewness.