Premium
Analysis of multivariate polychoric correlation models with incomplete data
Author(s) -
Lee SikYum,
Chiu YiuMing
Publication year - 1990
Publication title -
british journal of mathematical and statistical psychology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 3.157
H-Index - 51
eISSN - 2044-8317
pISSN - 0007-1102
DOI - 10.1111/j.2044-8317.1990.tb00931.x
Subject(s) - polychoric correlation , mathematics , multivariate statistics , statistics , restricted maximum likelihood , maximum likelihood , principal component analysis , multivariate normal distribution , multivariate analysis , correlation , missing data , econometrics , geometry
Several methods of analysing the multivariate polychoric correlation model with missing data are studied. They are the direct maximum likelihood method, the partition maximum likelihood method, the complete data method, the mean replacement method, the regression replacement method and the principal components method. Based on results of some simulation studies, the direct maximum likelihood method and the partition maximum likelihood method are recommended.