Premium
The analysis of patterned correlation matrices by generalized least squares
Author(s) -
Browne M. W.
Publication year - 1977
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.2044-8317.1977.tb00730.x
Subject(s) - mathematics , estimator , covariance matrix , generalized least squares , statistics , computation , population , covariance , least squares function approximation , matrix (chemical analysis) , correlation , algorithm , sociology , composite material , materials science , demography , geometry
Generalized least squares estimators of population parameters and an asymptotic chi‐square test are given for situations where sets of elements of a population correlation matrix are equal or have specified values. These estimators are asymptotically equivalent to maximum‐likelihood estimators but involve less computation. Corresponding formulae for the hypothesis that sets of elements of a population covariance matrix are equal or have specified values are given for comparative purposes. A numerical illustration is provided.