Abstract
The late Professor Yanai has contributed to many fields ranging from aptitude diagnostics, epidemiology, and nursing to psychometrics and statistics. This paper reviews some of his accomplishments in multivariate analysis through his collaborative work with the present author, along with some untold episodes for the inception of key ideas underlying the work. The various topics covered include constrained principal component analysis, extensions of Khatri’s lemma, theWedderburn-Guttman theorem, ridge operators, generalized constrained canonical correlation analysis, and causal inference. A common thread running through all of them is projectors and singular value decomposition, which are the main subject matters of a recent monograph by Yanai, Takeuchi, and Takane [60].
References
[1] U. Böckenholt, I. Böckenholt, Canonical analysis of contingency tables with linear constraints, Psychometrika, 55 (1990), 633–639. 10.1007/BF02294612Search in Google Scholar
[2] M. T. Chu, R. E. Funderlic, G. H. Golub, A rank-one reduction formula and its applications to matrix factorizations, SIAM Review, 37 (1995), 512–530. 10.1137/1037124Search in Google Scholar
[3] R. E. Cline, R. E. Funderlic, The rank of a difference of matrices and associated generalized inverses, Linear Algebra Appl., 24 (1979), 185–215. 10.1016/0024-3795(79)90158-7Search in Google Scholar
[4] W. G. Cochran, The distribution of quadratic forms in a normal system with applications to analysis of covariance, Proc. Camb. Phil. Soc., 30 (1934), 178–191. 10.1017/S0305004100016595Search in Google Scholar
[5] A. Galantai, A note on generalized rank reduction, Act. Math. Hung., 116 (2007), 239–246. 10.1007/s10474-007-6038-1Search in Google Scholar
[6] L. Guttman, General theory and methods for matric factoring, Psychometrika, 9 (1944), 1–16. 10.1007/BF02288709Search in Google Scholar
[7] L. Guttman, A necessary and sufficient formula for matric factoring, Psychometrika, 22 (1957), 79–81. 10.1007/BF02289212Search in Google Scholar
[8] F. R. Helmert, Adjustment Computations by the Method of Least Squares (in German), 2nd Edition (Teubnei, Leipzig, 1907). Search in Google Scholar
[9] A. E. Hoerl, R. W. Kennard, Ridge regression: Biased estimation for nonorthogonal problems, Technometrics, 12 (1970), 55–67. 10.1080/00401706.1970.10488634Search in Google Scholar
[10] L. Hubert, J. Meulman, W. J. Heiser, Two purposes of matrix factorization: A historical appraisal, SIAM Review, 42 (2000), 68–82. 10.1137/S0036144598340483Search in Google Scholar
[11] H. Hwang, Y. Takane, Generalized Structured Component Analysis: A Component-Based Approach to Structural Equation Modeling (Chapman and Hall/CRC Press, Boca Raton, FL., 2014). 10.1201/b17872Search in Google Scholar
[12] C. G. Khatri, A simplified approach to the derivation of the theorems on the rank of a matrix, J. of the Maharaja Sayajirao University of Baroda 10 (1961), 1–5. Search in Google Scholar
[13] C. G. Khatri, A note on a MANOVA model applied to problems in growth curve, Ann. I. Stat. Math., 18 (1966), 75–86. 10.1007/BF02869518Search in Google Scholar
[14] C. G. Khatri, Some properties of BLUE in a linear model and canonical correlations associated with linear transformations, J. Multivariate Anal., 34 (1990), 211–226. 10.1016/0047-259X(90)90036-HSearch in Google Scholar
[15] L. R. LaMotte, A direct derivation of the REML likelihood function, Stat. Pap., 48 (2007), 321–327. 10.1007/s00362-006-0335-6Search in Google Scholar
[16] R. J. Light, B. H. Margolin, An analysis of variance of categorical data, J. Am. Stat. Assoc., 66 (1971), 534–544. 10.1080/01621459.1971.10482297Search in Google Scholar
[17] S. Loisel, Y. Takane, Partitions of Pearson’s chi-square statistic for frequency tables: A comprehensive account, Computation. Stat., in press. Search in Google Scholar
[18] T. Ogasawara, M. Takahashi, Indepedence of quadratic forms in normal system, J. Sci. Hiroshima University, 15 (1951), 1–9. Search in Google Scholar
[19] K. Pearson, On the criterion that a given system of deviation from the probable in the case of a correlated system of variables is such that it can be reasonably supposed to have arisen from random sampling, Philos. Mag., 50 (1900), 157–172. 10.1080/14786440009463897Search in Google Scholar
[20] R. F. Potthoff, S. N. Roy, A generalized multivariate analysis of variance model useful for growth curve problem, Biometrika, 51 (1964), 313–326. 10.1093/biomet/51.3-4.313Search in Google Scholar
[21] S. Puntanen, G. P. H. Styan, J. Isotalo, Matrix Tricks for Linear Statistical Models (Springer, Berlin, 2011). 10.1007/978-3-642-10473-2Search in Google Scholar
[22] J. O. Ramsay, B. M. Silverman, Functional Data Analysis, Second Edition (Springer, New York, 2005). 10.1007/b98888Search in Google Scholar
[23] C. R. Rao, Linear Statistical Inference and Its Applications, Second Edition (Wiley, New York, 1973). 10.1002/9780470316436Search in Google Scholar
[24] C. R. Rao, S. K. Mitra, Generalized Inverse of Matrices amd Its Applications (Wiley, New York, 1971). Search in Google Scholar
[25] C. R. Rao, H. Yanai, General definition and decomposition of projectors and some applications to statistical problems, J. Stat. Plan. Infer., 3 (1979), 1–17. 10.1016/0378-3758(79)90038-7Search in Google Scholar
[26] B. Schaffrin, Personal communication (2015). Search in Google Scholar
[27] G. F. A. Seber, Multivariate Observations (Wiley, New York, 1984.) 10.1002/9780470316641Search in Google Scholar
[28] A. Shapiro, Asymptotic theory of overparameterized structural models, J. Am. Stat. Assoc., 81 (1986), 142–149. 10.1080/01621459.1986.10478251Search in Google Scholar
[29] Y. Takane, Relationships among various kinds of eigenvalue and singular value decompositions, In: H. Yanai, A. Okada, K. Shigemasu, Y. Kano, and J. Meulman (Eds.), New Developments in Psychometrics (Springer, Tokyo, 2003), 45–56. 10.1007/978-4-431-66996-8_4Search in Google Scholar
[30] Y. Takane, More on regularization and (generalized) ridge operators, In: K. Shigemasu, A. Okada, T. Imaizumi, T. Hoshino (Eds.), New Trends in Psychometrics (University Academic Press, Tokyo, 2008) 443–452. Search in Google Scholar
[31] Y. Takane, Constrained Principal Component Analysis and Related Techniques (Chapman and Hall/CRC Press, Boca Raton, FL, 2013). Search in Google Scholar
[32] Y. Takane, M. A. Hunter, Constrained principal component analysis: A comprehensive theory, Appl. Algebr. Eng. Comm., 12 (2001), 391–419. 10.1007/s002000100081Search in Google Scholar
[33] Y. Takane, M. A. Hunter, New family of constrained principal component analysis (CPCA), Linear Algebra Appl., 434 (2011), 2539–2555. 10.1016/j.laa.2011.01.002Search in Google Scholar
[34] Y. Takane, H. Hwang, Generalized constrained canonical correlation analysis, Multivar. Behav. Res., 37 (2002), 163–195. 10.1207/S15327906MBR3702_01Search in Google Scholar
[35] Y. Takane, H. Hwang, Regularized multiple correspondence analysis. In: J. Blasius, M. J. Greenacre (Eds.), Multiple correspondence analysis and related methods (Chapman and Hall, London, 2006) 259–279. 10.1201/9781420011319.ch11Search in Google Scholar
[36] Y. Takane, S. Jung, Regularized partial and/or constrained redundancy analysis, Psychometrika, 73 (2008), 671–690. 10.1007/s11336-008-9067-ySearch in Google Scholar
[37] Y. Takane, S. Jung, Regularized nonsymmetric correspondence analysis, Comput. Stat. Data An., 53 (2009), 3159–3170. 10.1016/j.csda.2008.09.004Search in Google Scholar
[38] Y. Takane, S. Jung, Tests of ignoring and eliminating in nonsymmetric correspondence analysis, Adv. Data Anal. Classif., 3 (2009), 315–340. 10.1007/s11634-009-0054-7Search in Google Scholar
[39] Y. Takane, T. Shibayama, Principal component analysis with external information on both subjects and variables, Psyhometrika, 56 (1991), 97–120. 10.1007/BF02294589Search in Google Scholar
[40] Y. Takane, H. Yanai, On oblique projectors, Linear Algebra Appl., 289 (1999), 297–310. 10.1016/S0024-3795(98)10180-5Search in Google Scholar
[41] Y. Takane, H. Yanai, On the Wedderburn-Guttman theorem, Linear Algebra Appl. 410 (2005), 267–278. 10.1016/j.laa.2005.08.009Search in Google Scholar
[42] Y. Takane, H. Yanai, On ridge operators, Linear Algebra Appl., 428 (2008), 1778–1790. 10.1016/j.laa.2007.10.017Search in Google Scholar
[43] Y. Takane, L. Zhou, On two expressions of the MLE for a special case of the extended growth curve models, Linear Algebra Appl., 436 (2012), 2567–2577. 10.1016/j.laa.2011.09.015Search in Google Scholar
[44] Y. Takane, L. Zhou, Anatomy of Pearson’s chi-square statistic in three-way contingency tables, In: R. E. Millsap, L. A. van der Ark, D. M. Bolt, C. M. Woods (Eds.), New Developments in Quantitative Psychology (Springer, New York, 2013), 41–57. 10.1007/978-1-4614-9348-8_4Search in Google Scholar
[45] Y. Takane, H. Hwang, H. Abdi, Regularized multiple-set canonical correlation analysis, Psychometrika, 73 (2008), 753–775. 10.1007/s11336-008-9065-0Search in Google Scholar
[46] Y. Takane, K. Jung, H. Hwang, Regularized growth curve models, Comput. Stat. Data An., 55 (2011), 1041–1052. 10.1016/j.csda.2010.08.012Search in Google Scholar
[47] Y. Takane, H. A. L. Kiers, J. de Leeuw, Component analysiswith different constraints on different dimensions, Psychometrika, 60 (1995), 259–280. 10.1007/BF02301416Search in Google Scholar
[48] Y. Takane, H. Yanai, H. Hwang, An improved method for generalized constrained canonical correlation analysis, Comp. Stat. Data An., 50 (2006), 221–241. 10.1016/j.csda.2004.07.016Search in Google Scholar
[49] Y. Takane, H. Yanai, S. Mayekawa, Relationships among several methods of linearly constrained correspondence analysis, Psychometrika, 56 (1991), 667–684. 10.1007/BF02294498Search in Google Scholar
[50] K. Takeuchi, H. Yanai, B. N.Mukherjee, The Foundation ofMultivariate Analysis (Wiley Eastern, New Delhi, and Halsted Press, New York, 1982). Search in Google Scholar
[51] C. J. F. ter Braak, Canonical correspondence analysis: A new eigenvector technique for multivariate direct gradient analysis, Ecology, 67 (1986), 1167–1179. 10.2307/1938672Search in Google Scholar
[52] Y. Tian, The Moore-Penrose inverses of m× n blockmatrices and their applications, Linear Algebra Appl., 283 (1998), 35–60. 10.1016/S0024-3795(98)10049-6Search in Google Scholar
[53] Y. Tian, Upper and lower bounds for ranks ofmatrix expressions using generalized inverses, Linear Algebra Appl., 355 (2002), 187–214. 10.1016/S0024-3795(02)00345-2Search in Google Scholar
[54] Y. Tian, G. P. H. Styan,Onsomematrix equalities for generalized inverseswith applications, Linear Algebra Appl., 430 (2009), 2716–2733. 10.1016/j.laa.2008.12.005Search in Google Scholar
[55] A. P. Verbyla, A conditional derivation of residual maximum likelihood, Aust. J. Stat., 32 (1990), 227–230. 10.1111/j.1467-842X.1990.tb01015.xSearch in Google Scholar
[56] J. H. M. Wedderburn, Lectures on Matrices, Colloquium Publication, Vol. 17 (American Mathematical Society, Providence, 1934). 10.1090/coll/017Search in Google Scholar
[57] H. Yanai, Factor analysis with external criteria, Jpn. Psychol. Res., 12 (1970), 143–153. 10.4992/psycholres1954.12.143Search in Google Scholar
[58] H. Yanai, Some generalized forms of least squares g-inverse, minimumnorm g-inverse and Moore-Penrose inversematrices, Comput. Stat. Data An., 10 (1990), 251–260. 10.1016/0167-9473(90)90005-3Search in Google Scholar
[59] H. Yanai, Y. Takane, Canonical correlation analysis with linear constraints, Linear Algebra Appl., 176 (1992), 75–82. 10.1016/0024-3795(92)90211-RSearch in Google Scholar
[60] H. Yanai, K. Takeuchi, Y. Takane, Projection Matrices, Generalized Inverse Matrices, and Singular Value Decomposition (Springer, New York, 2011). 10.1007/978-1-4419-9887-3Search in Google Scholar
©2016 Yoshio Takane
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.
