Exact and approximate representations for the sum Dirichlet process

被引:150
作者
Ishwaran, H [1 ]
Zarepour, M [1 ]
机构
[1] Cleveland Clin Fdn, Dept Biostat & Epidemiol, Cleveland, OH 44195 USA
来源
CANADIAN JOURNAL OF STATISTICS-REVUE CANADIENNE DE STATISTIQUE | 2002年 / 30卷 / 02期
关键词
almost sure truncation; finite dimensional Dirichlet prior; Levy measure; Poisson process; random probability measure; stick-breaking prior; weak convergence;
D O I
10.2307/3315951
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 process can be regarded as a random probability measure for which the authors examine various sum representations. They consider in particular the gamma process construction of Ferguson (1973) and the "stick-breaking" construction of Sethuraman (1994). They propose a Dirichlet finite sum representation that strongly approximates the Dirichlet process. They assess the accuracy of this approximation and characterize the posterior that this new prior leads to in the context of Bayesian nonparametric hierarchical models.
引用
收藏
页码:269 / 283
页数:15
相关论文
共 43 条