On the extension of Gauss-Markov theorem to complex multivariate linear models

  • J. N. Srivastava


The purpose of this paper is to develop a theory of linear estimation under various multivariate linear models, which are more general than the usual model to which the standard techniques of multivariate analysis of variance are applicable. In particular, necessary and sufficient conditions under which (unique) best linear unbiased estimates of linear functions of (location) parameters exist are obtained. An extension of the Gauss-Markov theorem to the standard multivariate model was first made by the author in [13]. In this paper, further generalizations of the result to multiresponse designs where the standard technique is inapplicable are considered.


Suffix Column Space Observation Matrix Multivariate Linear Model Linear Unbiased Estimate 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    T. W. Anderson,Introduction to Multivariate Analysis, John Wiley, New York, 1956.Google Scholar
  2. [2]
    R. C. Bose,Mimeographed Notes on Analysis of Variance, Institute of Statistics, University of North Carolina, 1958.Google Scholar
  3. [3]
    D. R. Cox,Planning of Experiments, John Wiley, New York, 1960.Google Scholar
  4. [4]
    A. T. James, “The relationship algebra of a PBIB design,”Ann. Math. Statist., 28 (1957), 349–361.CrossRefMathSciNetGoogle Scholar
  5. [5]
    O. Kempthorne,The design and Analysis of Experiments, John Wiley, New York, 1952.zbMATHGoogle Scholar
  6. [6]
    F. M. Lord, “Estimation of parameters from incomplete data,”J. Amer. Statist. Ass., 50 (1955), 870–876.zbMATHCrossRefMathSciNetGoogle Scholar
  7. [7]
    J. Ogawa and G. Ishii, “The relationship algebra and the analysis of variance of a PBIB design,”Ann. Math. Statist., 36 (1965), 1815–1828.zbMATHCrossRefMathSciNetGoogle Scholar
  8. [8]
    C. R. Rao, “Analysis of dispersion with incomplete observations on one of the characters,”J. Roy. Statist. Soc., Ser. B, 18 (1956), 259.zbMATHMathSciNetGoogle Scholar
  9. [9]
    S. N. Roy,Some Aspects of Multivariate Analysis, John Wiley, New York, 1957.Google Scholar
  10. [10]
    S. N. Roy and J. N. Srivastava,Hierarchical and p-block Multiresponse Designs and Their Analysis, Mahalanobis dedicatory volume, Indian Statistical Institute, Calcutta, 1962.Google Scholar
  11. [11]
    H. Scheffe,Analysis of Variance, John Wiley, New York, 1959.zbMATHGoogle Scholar
  12. [12]
    H. Smith, R. Gnanadesikan and J. B. Hughes, “Multivariate analysis of variance,”Biometrics, 18 (1962), 22–41.zbMATHCrossRefMathSciNetGoogle Scholar
  13. [13]
    J. N. Srivastava, “A multivariate extension of the Gauss-Markov theorem,”Ann. Inst. Statist. Math., 17 (1965), 63–66.zbMATHCrossRefMathSciNetGoogle Scholar
  14. [14]
    J. N. Srivastava, “Some generalizations of multivariate analysis of variance,”Multivariate Analysis (edited by P. R. Krishnaiah), Academic Press, New York, (1966), 129–148.Google Scholar
  15. [15]
    J. N. Srivastava, “On a general class of designs for multiresponse experiments,”N. C. Inst. of Statist., Mimeo Series No. 402, 1964.Google Scholar
  16. [16]
    J. N. Srivastava, “On the extension of Gauss-Markov theorem to complex multivariate linear models,” (abstract),Ann. Math. Statist., 37 (1966), 1860.Google Scholar
  17. [17]
    J. N. Srivastava, “Incomplete multiresponse designs,”Sankhyā A. 28 (1966), 377–388.zbMATHGoogle Scholar
  18. [18]
    I. M. Trawinski,Incomplete-Variable Designs, (unpublished thesis), V. P. I., Blacksburg, Virginia, 1961.Google Scholar
  19. [19]
    I. M. Trawinski and R. E. Bargmann, “Maximum likelihood estimation with incomplete multivariate data,”Ann. Math. Statist., 35 (1964), 647–657.zbMATHCrossRefMathSciNetGoogle Scholar

Copyright information

© The Institute of Statistical Mathematics 1967

Authors and Affiliations

  • J. N. Srivastava
    • 1
  1. 1.University of NebraskaNebraskaUSA

Personalised recommendations