Mixed beta regression: A Bayesian perspective

被引：85

作者：

Figueroa-Zuniga, Jorge I. ^{[1
]}

Arellano-Valle, Reinaldo B. ^{[2
]}

Ferrari, Silvia L. P. ^{[3
]}

机构：

[1] Univ Concepcion, Dept Stat, Santiago, Chile

[2] Pontificia Univ Catolica Chile, Dept Stat, Santiago, Chile

[3] Univ Sao Paulo, Dept Stat, BR-05508 Sao Paulo, Brazil

来源：

COMPUTATIONAL STATISTICS & DATA ANALYSIS | 2013年 / 61卷

关键词：

Bayesian analysis; Beta distribution; Beta regression; Continuous proportions; Mixed models; RUN LENGTH CONTROL; MODELS; SIMULATIONS; INFERENCE; VARIABLES;

D O I：

10.1016/j.csda.2012.12.002

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

This paper builds on recent research that focuses on regression modeling of continuous bounded data, such as proportions measured on a continuous scale. Specifically, it deals with beta regression models with mixed effects from a Bayesian approach. We use a suitable parameterization of the beta law in terms of its mean and a precision parameter, and allow both parameters to be modeled through regression structures that may involve fixed and random effects. Specification of prior distributions is discussed, computational implementation via Gibbs sampling is provided, and illustrative examples are presented. (C) 2012 Elsevier B.V. All rights reserved.

引用

页码：137 / 147

页数：11

共 24 条

[1]

Albert J.A., 2000, Generalized Linear Models: A Bayesian Perspective, DOI DOI 10.1201/9781482293456

[2] BAYESIAN-ANALYSIS OF BINARY AND POLYCHOTOMOUS RESPONSE DATA [J].