2013
DOI: 10.1016/j.csda.2012.12.002
|View full text |Cite
|
Sign up to set email alerts
|

Mixed beta regression: A Bayesian perspective

Abstract: 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, com… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

1
76
0
5

Year Published

2015
2015
2020
2020

Publication Types

Select...
6
3

Relationship

0
9

Authors

Journals

citations
Cited by 102 publications
(83 citation statements)
references
References 19 publications
1
76
0
5
Order By: Relevance
“…In [13], a Bayesian mixed-effects approach was proposed for the well known and widely applicable beta regression model [12]. Although the model is primarily suitable for response variables in the unit interval (0, 1), the authors suggested that the Bayesian beta mixed-effects regression model can be applied to variables bounded to any interval (a, b) by considering the transformation y * = (y − a)/(b − a).…”
Section: Model Formulationmentioning
confidence: 99%
“…In [13], a Bayesian mixed-effects approach was proposed for the well known and widely applicable beta regression model [12]. Although the model is primarily suitable for response variables in the unit interval (0, 1), the authors suggested that the Bayesian beta mixed-effects regression model can be applied to variables bounded to any interval (a, b) by considering the transformation y * = (y − a)/(b − a).…”
Section: Model Formulationmentioning
confidence: 99%
“…Zimprich (2010) added a random effect in the mean model to study reaction time of old people in a longitudinal survey. Further development of mixed beta regression models has been done by Figueroa et al (2013). They proposed mixed models for both mean and precision parameters and implemented Bayesian approach for parameter estimation.…”
Section: Introductionmentioning
confidence: 99%
“…Sob uma abordagem bayesiana, Bonat et al (2012) propõem modelos de regressão beta com efeitos aleatórios na estrutura de regressão de média e desenvolvem inferência estatística por meio da maximização da verossimilhança marginal e da metodologia bayesiana de algoritmo data cloning. Figueroa-Zúñiga et al (2012) estendem o modelo proposto por Ferrari e Cribari-Neto (2004) acrescentando efeitos aleatórios com distribuição normal e distribuição t-Student nas estruturas de regressão da média e precisão e discutem a especificação de distribuições a priori e a implementação computacional do modelo por meio do amostrador de Gibbs.…”
Section: 0unclassified
“…A classe de modelos de regressão beta com efeitos fixos tem sido estudada amplamente 1.1 por autores como Paolino (2001), Kieschnick e McCullough (2003), Ferrari e Cribari-Neto (2004), Smithson e Verkuilen (2006), Venezuela (2008), Simas et al (2010), entre outros. Porém, existem poucas publicações, até hoje, sobre a inclusão de efeitos aleatórios na classe de modelos beta e menos ainda considerando uma flexibilização da distribuição dos efeitos aleatórios (Figueroa-Zúñiga et al (2012)) e nenhum trabalho sobre a predição dos efeitos aleatórios, os métodos de seleção dos modelos e os diagnósticos dos modelos.…”
Section: Introductionunclassified