A test for additional information in canonical correlation analysis

DOI10.1007/BF02481050MaRDI QIDQ796935zbMATH OpenFDO

Authors Yasunori Fujikoshi

Publication date 1982

Published in Annals of the Institute of Statistical Mathematics (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1007/BF02481050

zbMATH Keywords

additional informationasymptotic expansioncanonical correlation analysisdiscriminant analysislikelihood ratio statisticnull distribution

Mathematics Subject Classification ID

Multivariate distribution of statistics (62H10) Hypothesis testing in multivariate analysis (62H15) Measures of association (correlation, canonical correlation, etc.) (62H20) Classification and discrimination; cluster analysis (statistical aspects) (62H30)

Review text

Étant donné quatre vecteurs aléatoires réels \(X_ 1,X_ 2,Y_ 1\) et \(Y_ 2\), l'A. détermine d'abord diverses conditions équivalentes pour que les analyses canoniques linéaires de \(X=(X_ 1,X_ 2)\) et \(Y=(Y_ 1,Y_ 2)\) d'une part, et, de \(X_ 1\) et \(Y_ 1\) d'autre part fournissent les mêmes coefficients canoniques non nuls, ce que l'on peut énoncer en disant \((H_ 1):(X_ 2,Y_ 2)\) n'apporte pas d'informations supplémentaires sur les relations entre \(X\) et \(Y\) à celles données par \((X_ 1,Y_ 1).\) Dans le cas où le couple \((X,Y)\) est gaussien, il détermine la loi sous l'hypothèse \((H_ 1)\) de la statistique du test de rapport de vraisemblance et en donne une approximation asymptotique. Des problèmes similaires sont ensuite envisagés dans le cas de l'analyse discriminante.

Recommendations

• Asymptotic study of canonical analysis
• Common canonical variates
• THE NON-NULL DISTRIBUTION OF THE LIKELIHOOD RATIO CRITERION FOR ADDITIONAL INFORMATION HYPOTHESIS IN CANONICAL CORRELATION ANALYSIS
• scientific article; zbMATH DE number 4064281
• Measures of Asociation between vectors

Cites work

• A GENERAL DISTRIBUTION THEORY FOR A CLASS OF LIKELIHOOD CRITERIA
• scientific article; zbMATH DE number 3136275 (Why is no real title available?)
• scientific article; zbMATH DE number 3604203 (Why is no real title available?)
• scientific article; zbMATH DE number 3335601 (Why is no real title available?)
• scientific article; zbMATH DE number 3373230 (Why is no real title available?)
• scientific article; zbMATH DE number 3090534 (Why is no real title available?)
• Linear Statistical Inference and its Applications
• RELATIONS BETWEEN TWO SETS OF VARIATES
• Simultaneous procedures for variable selection in multiple discriminant analysis
• TESTS OF SIGNIFICANCE IN MULTIVARIATE ANALYSIS
• THE STATISTICAL UTILIZATION OF MULTIPLE MEASUREMENTS

Cited in

(27)

• Asymptotic expansions of test statistics for dimensionality and additional information in canonical correlation analysis when the dimension is large
• Asymptotic distribution of the increase of the largest canonical correlation when one of the vectors is augmented
• Canonical correlation analysis with linear constraints
• Canonical analysis of two Euclidean subspaces and its applications
• Linear relative canonical analysis of Euclidean random variables, asymptotic study and some applications
• A study of redundancy of some variables in covariate discriminant analysis
• On tests for selection of variables and independence under multivariate regression models
• Inference for the invariance of canonical analysis under linear transformations.
• Jackknife bias correction of the AIC for selecting variables in canonical correlation analysis under model misspecification
• Strong consistency of log-likelihood-based information criterion in high-dimensional canonical correlation analysis
• A consistent variable selection method in high-dimensional canonical discriminant analysis
• Variable selection in discriminant analysis for mixed continuous-binary variables and several groups
• On invariance in canonical analysis with applications to covariate discriminant analysis
• High-dimensional AICs for selection of variables in discriminant analysis
• Asymptotic study of canonical analysis
• A variable selection criterion for two sets of principal component scores in principal canonical correlation analysis
• Understanding Canonical Correlation through the General Linear Model and Principal Components
• Study of redundancy of vector variables in canonical correlations
• scientific article; zbMATH DE number 4064281 (Why is no real title available?)
• scientific article; zbMATH DE number 472970 (Why is no real title available?)
• Computable error bounds for high-dimensional approximations of an LR statistic for additional information in canonical correlation analysis
• A Closed Testing Procedure for Canonical Correlations
• THE NON-NULL DISTRIBUTION OF THE LIKELIHOOD RATIO CRITERION FOR ADDITIONAL INFORMATION HYPOTHESIS IN CANONICAL CORRELATION ANALYSIS
• Measures of Asociation between vectors
• Asymptotic null and non-null distributions of test statistics for redundancy in high-dimensional canonical correlation analysis
• A mutual information criterion with applications to canonical correlation analysis and graphical models
• Direct variable selection for discrimination among several groups

This page was built for publication: A test for additional information in canonical correlation analysis

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q796935)