Premium
MULTIVARLYTE ANALYSIS OF CLUSTERED SURVEY DATA: TESTING EQUALITY OF COVARLANCE MATRICES
Author(s) -
Pezvaiz Muhammad Khalid,
Skinner C.J.
Publication year - 1992
Publication title -
australian journal of statistics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.434
H-Index - 41
eISSN - 1467-842X
pISSN - 0004-9581
DOI - 10.1111/j.1467-842x.1992.tb01362.x
Subject(s) - mathematics , sampling (signal processing) , transformation (genetics) , linearization , statistics , population , sample (material) , covariance , computer science , nonlinear system , biochemistry , chemistry , physics , demography , filter (signal processing) , chromatography , quantum mechanics , sociology , computer vision , gene
Summary The problem of testing the hypothesis of equality of covariance matrices in the presence of two‐stage sampling is considered. Asymptotic test procedures based on linearization, grouping and jackknifing with or without transformation are proposed. The finite sample properties of these procedures are investigated in sampling experiments both from simulated known distributions and from a natural population.