Efficient estimation for incomplete multivariate data

被引:1
|
作者
Jorgensen, Bent [1 ]
Petersen, Hans Chr. [1 ]
机构
[1] Univ So Denmark, Dept Math & Comp Sci, DK-5230 Odense M, Denmark
关键词
EM algorithm; Estimating function; Fisher scoring algorithm; Godambe information matrix; Missing data; REML estimation; MAXIMUM-LIKELIHOOD; INFORMATION MATRIX; REGRESSION-MODELS; MISSING VALUES; PARAMETERS;
D O I
10.1016/j.jspi.2011.11.024
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
We review the Fisher scoring and EM algorithms for incomplete multivariate data from an estimating function point of view, and examine the corresponding quasi-score functions under second-moment assumptions. A bias-corrected REML-type estimator for the covariance matrix is derived, and the Fisher, Godambe and empirical sandwich information matrices are compared. We make a numerical investigation of the two algorithms, and compare with a hybrid algorithm, where Fisher scoring is used for the mean vector and the EM algorithm for the covariance matrix. (C) 2011 Elsevier B.V. All rights reserved.
引用
收藏
页码:1215 / 1224
页数:10
相关论文
共 50 条