A hybrid EM/Gauss-Newton algorithm for maximum likelihood in mixture distributions

被引:29
作者
Aitkin, M
Aitkin, I
机构
[1] UNIV WESTERN AUSTRALIA,DEPT MATH,NEWCASTLE TYNE,TYNE & WEAR,ENGLAND
[2] UNIV NEWCASTLE UPON TYNE,DEPT STAT,NEWCASTLE TYNE,TYNE & WEAR,ENGLAND
[3] CURTIN UNIV TECHNOL,SCH PUBL HLTH,DEPT EPIDEMIOL & BIOSTAT,BENTLEY,WA 6102,AUSTRALIA
关键词
mixtures; maximum likelihood; Gauss-Newton; EM algorithm;
D O I
10.1007/BF00162523
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
A faster alternative to the EM algorithm in finite mixture distributions is described, which alternates EM iterations with Gauss-Newton iterations using the observed information matrix. At the expense of modest additional analytical effort in obtaining the observed information, the hybrid algorithm reduces the computing time required and provides asymptotic standard errors at convergence. The algorithm is illustrated on the two-component normal mixture.
引用
收藏
页码:127 / 130
页数:4
相关论文
共 9 条
[1]  
AITKIN M, 1980, COMPSTAT 1980
[2]   MAXIMUM LIKELIHOOD FROM INCOMPLETE DATA VIA EM ALGORITHM [J].
DEMPSTER, AP ;
LAIRD, NM ;
RUBIN, DB .
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-METHODOLOGICAL, 1977, 39 (01) :1-38
[3]   MAXIMUM-LIKELIHOOD ESTIMATION OF THE PARAMETERS IN A MIXTURE OF 2 UNIVARIATE NORMAL-DISTRIBUTIONS - A COMPARISON OF DIFFERENT ALGORITHMS [J].
EVERITT, BS .
STATISTICIAN, 1984, 33 (02) :205-215
[4]  
LOUIS TA, 1982, J ROY STAT SOC B MET, V44, P226
[5]  
McLachlan G.J., 1988, Mixture models: inference and applications to clustering, V38
[6]  
MEILIJSON I, 1989, J ROY STAT SOC B MET, V51, P127
[7]   USING EM TO OBTAIN ASYMPTOTIC VARIANCE - COVARIANCE MATRICES - THE SEM ALGORITHM [J].
MENG, XL ;
RUBIN, DB .
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1991, 86 (416) :899-909
[8]   MIXTURE DENSITIES, MAXIMUM-LIKELIHOOD AND THE EM ALGORITHM [J].
REDNER, RA ;
WALKER, HF .
SIAM REVIEW, 1984, 26 (02) :195-237
[9]  
TITTERINGTON DM, 1985, STAT ANAL FINITE MIX