MIXTURE-MODELS FOR MATCHED PAIRS

被引：2

作者：

LESPERANCE, ML

KALBFLEISCH, JD

机构：

[1] UNIV VICTORIA,DEPT MATH & STAT,VICTORIA V8W 3P4,BC,CANADA

[2] UNIV WATERLOO,DEPT STAT & ACTUARIAL SCI,WATERLOO N2L 3G1,ONTARIO,CANADA

来源：

CANADIAN JOURNAL OF STATISTICS-REVUE CANADIENNE DE STATISTIQUE | 1994年 / 22卷 / 01期

关键词：

CONDITIONAL LIKELIHOOD; CORRELATED BINARY REGRESSION; SEMIPARAMETRIC PROFILE LIKELIHOOD; MATCHED PAIRS; MIXTURE MODELS; SEMIPARAMETRIC MIXTURE MODELS;

D O I：

10.2307/3315823

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：

Mixture models for matched pairs arise when the pair-specific parameter is assumed to be a random quantity. We explore the use of semiparametric maximum-likelihood methods for a family of mixture models for matched pairs. The geometry of mixture likelihoods provides insight into the properties of these models.

引用

页码：65 / 74

页数：10

共 13 条

[1]

ALTHAM PME, 1971, BIOMETRIKA, V58, P561, DOI 10.1093/biomet/58.3.561

[2]

Cox DR, 1989, ANAL BINARY DATA, V2nd

[3] CONSISTENT ESTIMATION IN THE RASCH MODEL BASED ON NONPARAMETRIC MARGINS [J].