Advertisement

Statistical Papers

, Volume 36, Issue 1, pp 287–298 | Cite as

Matrix-trace Cauchy-Schwarz inequalities and applications in canonical correlation analysis

  • Shuangzhe Liu
  • Heinz Neudecker
Articles

Abstract

Various matrix-trace Cauchy-Schwarz and related inequalities involving positive semidefinite matrices are obtained. Applications of some of these results to canonical correlation analysis are presented.

Key words

Matrix trace positive semidefinite matrix Cauchy-Schwarz inequality canonical correlation 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Anderson, T.W. [1984] An Introduction to Multivariate Statistical Analysis, 2nd edn.Wiley, New York.zbMATHGoogle Scholar
  2. Chaganty, N.R. [1993] Comment on “A matrix formulation on how deviant an observation can be”, by I. Olkin.The Amer. Statist., 47, 158.Google Scholar
  3. Giri, N. [1977] Multivariate Statistical Inference.Academic Press, New York.zbMATHGoogle Scholar
  4. Kshirsagar, A.M. [1972] Multivariate Analysis.Dekker, New York.zbMATHGoogle Scholar
  5. Leon, S.J. [1994] Maximizing bilinear forms subject to linear constraints.Linear Algebra Appl., 210, 49–58.MathSciNetCrossRefzbMATHGoogle Scholar
  6. Lin, C.T. [1990] Extremes of determinants and optimality of canonical variables.Commun. Statist. Simul. Comp., 19, 1415–1430.MathSciNetCrossRefzbMATHGoogle Scholar
  7. Magnus, J.R. and Neudecker, H. [1991] Matrix Differential Calculus with Applications in Statistics and Econometrics.Wiley, Chichester.zbMATHGoogle Scholar
  8. Mardia, K.V., Kent, J.T. and Bibby, J.M. [1979] Multivariate Analysis.Academic Press, New York.zbMATHGoogle Scholar
  9. Muirhead, R.J. [1982] Aspects of Multivariate Statistical Theory.Wiley, New York.CrossRefzbMATHGoogle Scholar
  10. Neudecker, H. and Liu, S. [1994] Comment on “A matrix formulation on how deviant an observation can be”, by I. Olkin.The Amer. Statist., forthcoming.Google Scholar
  11. Olkin, I. [1992] A matrix formulation on how deviant an observation can be.The Amer. Statist., 46, 205–209.MathSciNetGoogle Scholar
  12. Press, S.J. [1972] Applied Multivariate Analysis.Holt, Rinehart and Winston, Inc., New York.zbMATHGoogle Scholar
  13. Rao, C.R. [1973] Linear Statistical Inference and Its Applications, 2nd edn.Wiley, New York.CrossRefzbMATHGoogle Scholar
  14. Rao, C.R. [1981] A lemma on g-inverse of a matrix and computation of correlation coefficients in the singular case.Commun. Statist. Theor. Meth., 10, 1–10.MathSciNetCrossRefzbMATHGoogle Scholar
  15. Toutenburg, H. [1992] Lineare Modelle.Physica-Verlag, Heidelberg.CrossRefGoogle Scholar
  16. Trenkler, G. [1994] Comment on “A matrix formulation on how deviant an observation can be”, by I.Olkin.The Amer. Statist., 48, 60–61.CrossRefGoogle Scholar
  17. Yanai, H. and Takane, Y. [1992] Canonical correlation analysis with linear constraints.Linear Algebra Appl., 176, 75–89.MathSciNetCrossRefzbMATHGoogle Scholar

Copyright information

© Springer-Verlag 1995

Authors and Affiliations

  • Shuangzhe Liu
    • 1
  • Heinz Neudecker
    • 1
  1. 1.Institute of Actuarial Science and EconometricsUniversity of AmsterdamAmsterdamThe Netherlands

Personalised recommendations