THE EM ALGORITHM FOR GRAPHICAL ASSOCIATION MODELS WITH MISSING DATA

被引:476
|
作者
LAURITZEN, SL [1 ]
机构
[1] AALBORG UNIV,INST ELECTR SYST,DEPT MATH & COMP SCI,DK-9220 AALBORG O,DENMARK
关键词
CONTINGENCY TABLES; BELIEF NETWORKS; DECOMPOSABLE MODELS; EXPERT SYSTEMS; HIERARCHICAL MODELS; LATENT VARIABLES; PROBABILITY PROPAGATION; RECURSIVE MODELS;
D O I
10.1016/0167-9473(93)E0056-A
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
It is shown how the computational scheme of Lauritzen and Spiegelhalter (1988) can be exploited to perform the E-step of the EM algorithm when applied to finding maximum likelihood estimates or penalized maximum likelihood estimates in hierarchical log-linear models and recursive models for contingency tables with missing data. The generalization to mixed association models introduced in Lauritzen and Wermuth (1989) and Edwards (1990) is indicated.
引用
收藏
页码:191 / 201
页数:11
相关论文
共 50 条