Email this article Canonical Correlation Analysis of Principal Component Scores for Multiple-set Random Vectors
Abstract
Canonical correlation analysis (CCA) is often used to analyze correlations between the variables of two random vectors. As an extension of CCA, multiple-set canonical correlation analysis (MCCA) was proposed to analyze correlations between multiple-set random vectors. However, sometimes interpreting MCCA results may not be as straightforward as interpreting CCA results. Principal CCA (PCCA), which uses CCA between two sets of principal component (PC) scores, was proposed to address these difficulties in CCA. We propose multiple-set PCCA (MPCCA) by applying the idea to multiple-set of PC scores. PCs are ranked in descending order according to the amount of information they contain. Therefore, it is enough to use only a few PC scores from the top instead of using all PC scores. Decreasing the number of PC makes it easy to interpret the result. We confirmed the effectiveness of MPCCA using simulation studies and a practical example.
References
@article{Anderson51,
author = {Anderson, T W.},
journal = {The Annals of Mathematical Statistics},
number = {3},
pages = {327--351},
title = {Estimating Linear Restrictions on Regression Coefficients for Multivariate Normal Distributions},
volume = {22},
year = {1951},
}
@article{Anderson63,
author = {Anderson, T W.},
journal = {The Annals of Mathematical Statistics},
number = {1},
pages = {122--148},
title = {Asymptotic Theory for Principal Component Analysis},
volume = {34},
year = {1963},
}
@book{Anderson03,
title = {An Introduction to Multivariate Statistical Analysis, 3rd Edition},
author = {Anderson, T W.},
publisher = {Wiley},
year = {2003},
address = {New York},
}
@article{Hotelling33,
author = {Hotelling, H.},
journal = {Journal of Educational Psychology},
number = {6},
pages = {417--441},
title = {Analysis of a Complex of Statistical Variables into Principal Components},
volume = {24},
year = {1933},
}
@article{Hotelling35,
author = {Hotelling, H.},
journal = {Journal of Educational Psychology},
number = {2},
pages = {139--142},
title = {The Most Predictable Criterion},
volume = {26},
year = {1935},
}
@article{Hotelling36,
author = {Hotelling, H.},
journal = {Biometrika},
number = {3},
pages = {321--377},
title = {Relations between Two Sets of Variates},
volume = {28},
year = {1936},
}
@article{Hwang13,
author = {Hwang, H. and Jung, K. and Takane, Y. and Woodward, T S.},
journal = {British Journal of Mathematical and Statistical Psychology},
number = {2},
pages = {308--321},
title = {A Unified Approach to Multiple-set Canonical Correlation Analysis and Principal Components Analysis},
volume = {66},
year = {2013},
}
@article{Kettenring71,
author = {Kettenring, J R.},
journal = {Biometrika},
number = {3},
pages = {433--451},
title = {Canonical Analysis of Several Sets of Variables},
volume = {58},
year = {1971},
}
@article{Lawley59,
author = {Lawley, D N.},
journal = {Biometrika},
number = {1--2},
pages = {59--66},
title = {Tests of Significance in Canonical Analysis},
volume = {46},
year = {1959},
}
@article{Ogura10,
author = {Ogura, T.},
journal = {Computational Statistics & Data Analysis},
number = {4},
pages = {1117--1123},
title = {A Variable Selection Method in Principal Canonical Correlation Analysis},
volume = {54},
year = {2010},
}
@article{Ogura13,
author = {Ogura, T. and Fujikoshi, Y. and Sugiyama, T.},
journal = {Communications in Statistics - Theory and Methods},
number = {12},
pages = {2118--2135},
title = {A Variable Selection Criterion for Two Sets of Principal Component Scores in Principal Canonical Correlation Analysis},
volume = {42},
year = {2013},
}
@article{Palatella10,
author = {Palatella, L. and Miglietta, M M. and Paradisi, P. and Lionello, P.},
journal = {Natural Hazards and Earth System Sciences},
number = {4},
pages = {1647--1661},
title = {Climate Change Assessment for Mediterranean Agricultural Areas by Statistical Downscaling},
volume = {10},
year = {2010},
}
@article{Skourkeas13,
author = {Skourkeas, A. and Kolyva-Machera, F. and Maheras, P.},
journal = {Environmental and Ecological Statistics},
number = {3},
pages = {445--465},
title = {Improved Statistical Downscaling Models based on Canonical Correlation Analysis, for Generating Temperature Scenarios over Greece},
volume = {20},
year = {2013},
}
@article{Sugiyama07,
author = {Sugiyama, T. and Ogura, T. and Sakaori, F. and Yamada, T.},
journal = {Journal of the Japan Statistical Society},
number = {2},
pages = {239--251},
title = {Distributions and Bootstrap Method of Some Statistics in Principal Canonical Correlation Analysis},
volume = {37},
year = {2007},
}
@article{Tenenhaus11,
author = {Tenenhaus, A. and Tenenhaus, M.},
journal = {Psychometrika},
number = {2},
pages = {257--284},
title = {Regularized Generalized Canonical Correlation Analysis},
volume = {76},
year = {2011},
}
@article{TerBraak90,
author = {Ter Braak, C J.},
journal = {Psychometrika},
number = {3},
pages = {519--531},
title = {Interpreting Canonical Correlation Analysis through Biplots of Structure Correlations and Weights},
volume = {55},
year = {1990},
}
@book{Wolfram16,
title = {Mathematica},
author = {Wolfram Research, Inc.},
publisher = {Wolfram Research, Inc.},
year = {2016},
address = {Champaign, Illinois},
edition = {Version 11.0},
}
@article{Yamamoto07,
author = {Yamamoto, M. and Sugiyama, T. and Murakami, H. and Sakaori, F.},
journal = {Computational Statistics & Data Analysis},
number = {9},
pages = {4707--4716},
title = {Correlation Analysis of Principal Components from Two Populations},
volume = {51},
year = {2007},
}
Full Text: pdf
