EFFICIENT COMPUTATIONAL PROCEDURES FOR THE ESTIMATION OF PARAMETERS IN MULTILEVEL MODELS BASED ON ITERATIVE GENERALIZED LEAST-SQUARES

被引：29

作者：

GOLDSTEIN, H ^{[1
]}

RASBASH, J ^{[1
]}

机构：

[1] UNIV LONDON,INST EDUC,LONDON WC1H 0AL,ENGLAND

来源：

COMPUTATIONAL STATISTICS & DATA ANALYSIS | 1992年 / 13卷 / 01期

基金：

英国经济与社会研究理事会;

关键词：

COMPUTATIONAL ALGORITHMS; ITERATIVE GENERALIZED LEAST SQUARES; HIERARCHICAL DATA; MAXIMUM LIKELIHOOD; MULTILEVEL MODELS; STATISTICAL COMPUTING;

D O I：

10.1016/0167-9473(92)90154-8

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

Computational procedures are described for the efficient estimation of the fixed and random parameters in multilevel random coefficient linear models. The estimation procedure is based on iterative generalised least squares. A number of special cases are described and the procedure is implemented within a general purpose statistical package.

引用

页码：63 / 71

页数：9

共 10 条

[1]

GOLDSTEIN H, 1989, BIOMETRIKA, V76, P622

[2]

GOLDSTEIN H, 1986, BIOMETRIKA, V73, P43

[3]

Goldstein H., 1987, MULTILEVEL MODELS ED

[4]

HEALY MJR, 1968, J R STAT SOC C-APPL, V17, P198

[5]

HEALY MJR, 1989, NANOSTAT USER MANUAL

[6] ON DERIVING THE INVERSE OF A SUM OF MATRICES [J].

HENDERSON, HV ;

SEARLE, SR .

SIAM REVIEW, 1981, 23 (01) :53-60

[7]

LONGFORD NT, 1987, BIOMETRIKA, V74, P812

[8]

PROSSER R, 1990, ML3 SOFTWARE 3 LEVEL

[9] A HIERARCHICAL MODEL FOR STUDYING SCHOOL EFFECTS [J].

RAUDENBUSH, S ;

BRYK, AS .

SOCIOLOGY OF EDUCATION, 1986, 59 (01) :1-17

[10]

Searle, 1982, MATRIX ALGEBRA USEFU

← 1 →