What Does Objective Mean in a Dirichlet-multinomial Process?

被引：17

作者：

Alvares, Danilo ^{[1
]}

Armero, Carmen ^{[1
]}

Forte, Anabel ^{[1
]}

机构：

[1] Univ Valencia, Valencia, Spain

来源：

INTERNATIONAL STATISTICAL REVIEW | 2018年 / 86卷 / 01期

关键词：

Bayesian inference; non-informative priors; posterior mean; DISTRIBUTIONS; MIXTURES; PRIORS;

D O I：

10.1111/insr.12231

中图分类号：

O21 [概率论与数理统计]; C8 [统计学];

学科分类号：

020208 ; 070103 ; 0714 ;

摘要：

The Dirichlet-multinomial process can be seen as the generalisation of the binomial model with beta prior distribution when the number of categories is larger than two. In such a scenario, setting informative prior distributions when the number of categories is great becomes difficult, so the need for an objective approach arises. However, what does objective mean in the Dirichlet-multinomial process? To deal with this question, we study the sensitivity of the posterior distribution to the choice of an objective Dirichlet prior from those presented in the available literature. We illustrate the impact of the selection of the prior distribution in several scenarios and discuss the most sensible ones.

引用

页码：106 / 118

页数：13

共 37 条

[1]

Agresti A., 2003, CATEGORICAL DATA ANA

[2] MIXTURES OF DIRICHLET DISTRIBUTIONS AND ESTIMATION IN CONTINGENCY-TABLES [J].