Nonparametric estimation in semi-parametric univariate mixture models

被引：13

作者：

Cruz-Medina, IR ^{[1
]}

Hettmansperger, TP ^{[1
]}

机构：

[1] Penn State Univ, Eberly Coll Sci, Dept Stat, University Pk, PA 16802 USA

来源：

JOURNAL OF STATISTICAL COMPUTATION AND SIMULATION | 2004年 / 74卷 / 07期

基金：

美国国家科学基金会;

关键词：

EM algorithm; finite mixture models; training samples;

D O I：

10.1080/00949650310001602158

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 [计算机应用技术]; 0835 [软件工程];

摘要：

We introduce new estimates of the mixing proportions, locations, and variances of the components of a finite univariate mixture model. We assume that the components are symmetric and differ only in the locations. No parametric model assumptions are imposed on the components. Further, when there is additional information available in the form of training samples that contain information concerning the mixing proportion, the new methods are robust to the symmetry assumption.

引用

页码：513 / 524

页数：12

共 12 条

[1]

CRUZMEDINA IR, 2002, UNPUB SEMIPARAMETRIC

[2]

CRUZMEDINA IR, 2001, THESIS PENN STATE U

[3]

HALL P, 1981, J ROY STAT SOC B MET, V43, P147

[4]

Hall P, 2003, ANN STAT, V31, P201

[5]

HALL P, 1984, J ROY STAT SOC B MET, V46, P465

[6]

Almost nonparametric inference for repeated measures in mixture models [J].

Hettmansperger, TP ;

Thomas, H .

JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY, 2000, 62 :811-825

[7]

COMPARISON OF ITERATIVE MAXIMUM LIKELIHOOD ESTIMATES OF PARAMETERS OF A MIXTURE OF 2 NORMAL DISTRIBUTIONS UNDER 3 DIFFERENT TYPES OF SAMPLE [J].

HOSMER, DW .

BIOMETRICS, 1973, 29 (04) :761-770

[8]

HUNTER W, 2002, UNPUB SEMIPARAMETRIC

[9]

MCLACHLAN G., 2000, WILEY SER PROB STAT, DOI 10.1002/0471721182

[10]

McLachlan G. J., 1997, EM ALGORITHM EXTENSI

← 1 2 →