Skip to main content
SpringerLink
Log in

Quadratic estimators of covariance components in a multivariate mixed linear model

  • Original Article
  • Published:
Statistical Methods and Applications Aims and scope Submit manuscript

Abstract

It is known that the Henderson Method III (Biometrics 9:226–252, 1953) is of special interest for the mixed linear models because the estimators of the variance components are unaffected by the parameters of the fixed factor (or factors). This article deals with generalizations and minor extensions of the results obtained for the univariate linear models. A MANOVA mixed model is presented in a convenient form and the covariance components estimators are given on finite dimensional linear spaces. The results use both the usual parametric representations and the coordinate-free approach of Kruskal (Ann Math Statist 39:70–75, 1968) and Eaton (Ann Math Statist 41:528–538, 1970). The normal equations are generalized and it is given a necessary and sufficient condition for the existence of quadratic unbiased estimators for covariance components in the considered model.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  • Baksalary JK, Kala R (1976) Criteria for estimability in multivariate linear models. Math Operationsforsch Statist Ser Statistics 7:5–9

    MathSciNet  Google Scholar 

  • Bates MD, DebRoy S (2004) Linear mixed models and penalized least squares. J Multivariate Anal 91:1–12

    Article  MATH  MathSciNet  Google Scholar 

  • Beganu G (1987a) Estimation of regression parameters in a covariance linear model. Stud Cerc Mat 39:3–10

    MATH  MathSciNet  Google Scholar 

  • Beganu G (1987b) Estimation of covariance components in linear models. A coordinate- free approach. Stud Cerc Mat 39:228–233

    MATH  MathSciNet  Google Scholar 

  • Beganu G (1992) A model of multivariate analysis of variance with applications to medicine. Econom Comput Econom Cybernet Stud Res 27:35–40

    MathSciNet  Google Scholar 

  • Biorn E (2004) Regression systems for unbalanced panel data: a stepwise maximum likelihood procedure. J Econometrics 122:281–295

    Article  MathSciNet  Google Scholar 

  • Calvin JA, Dykstra RL (1991) Maximum likelihood estimation of a set of covariance matrices under Lőwner order restrictions with applications to balanced multivariate variance components models. Ann Statist 19:850–869

    MATH  MathSciNet  Google Scholar 

  • Cossette H, Luong A (2003) Generalized least squares estimators for covariance parameters for credibility regression models with moving average errors. Insurance Math Econ 32:281–291

    Article  MATH  MathSciNet  Google Scholar 

  • Eaton ML (1970) Gauss–Markov estimation for multivariate linear models: a coordinate-free approach. Ann Math Statist 41:528–538

    MathSciNet  Google Scholar 

  • Halmos PR (1957) Finite dimensional vector spaces. Princeton, New Jersey

    Google Scholar 

  • Henderson CR (1953) Estimation of variance and covariance components. Biometrics 9:226–252

    Article  MathSciNet  Google Scholar 

  • Hultquist R, Atzinger EM (1972) The mixed effects model and simultaneous diagonalization of symmetric matrices. Ann Math Statist 43:2024–2030

    MathSciNet  Google Scholar 

  • Kleffe J (1977) Invariant methods for estimating variance components in mixed linear models. Math Operationsforsch Statist Ser Statistics 8:233–250

    MATH  MathSciNet  Google Scholar 

  • Klonecki W, Zontek S (1992) Admissible estimators of variance components obtained via submodels. Ann Statist 20:1454–1467

    MATH  MathSciNet  Google Scholar 

  • Kruskal W (1968) When are Gauss–Markov and least squares estimators identical? A coordinate-free approach. Ann Math Statist 39:70–75

    MathSciNet  Google Scholar 

  • Magnus JR, Neudecker H (1979) The commutation matrix: some properties and applications. Ann Statist 7:381–394

    MATH  MathSciNet  Google Scholar 

  • Milliken GA (1971) New criteria for estimability of linear models. Ann Math Statist 42:1588–1594

    MathSciNet  Google Scholar 

  • Neudecker H (1990) The variance matrix of a matrix quadratic form under normality assumptions. A derivation based on its moment-generating function. Math Operationsforsch Statist Ser Statist 3:455–459

    MathSciNet  Google Scholar 

  • Olsen A, Seely J, Birkes D (1976) Invariant quadratic unbiased estimation for two variance components. Ann Statist 4:878–890

    MATH  MathSciNet  Google Scholar 

  • Rao CR (1971a) Estimation of variance covariance components-MINQUE theory. J Multivariate Anal 1:257–276

    Article  Google Scholar 

  • Rao CR (1971b) Minimum variance quadratic unbiased estimation of variance components. J Multivariate Anal 1:445–456

    Article  Google Scholar 

  • Rao CR (1973) Linear statistical inference and its application. Wiley, New York

    Google Scholar 

  • Rao CR (1976) Estimation of parameters in a linear model. Ann Statist 4:1023–1037

    MATH  MathSciNet  Google Scholar 

  • Rao CR, Kleffe J (1988) Estimation of variance components and applications. North-Holland, Amsterdam

    MATH  Google Scholar 

  • Robinson GK (1991) That BLUP is a good think: the estimation of random effects. Stat Sci 6:15–51

    MATH  Google Scholar 

  • Searle SR (1971) Topics in variance components estimation. Biometrics 27:1–76

    Article  MathSciNet  Google Scholar 

  • Seely J (1970a) Linear spaces and unbiased estimation. Ann Math Statist 41:1725–1734

    MathSciNet  Google Scholar 

  • Seely J (1970b) Linear spaces and unbiased estimation. Application to the mixed linear model. Ann Math Statist 41:1735–1748

    MathSciNet  Google Scholar 

  • Seely J, Zyskind G (1971) Linear spaces and minimum variance unbiased estimation. Ann Math Statist 42:691–703

    MathSciNet  Google Scholar 

  • Watson GS (1967) Linear least squares regression. Ann Math Statist 38:1679–1699

    MathSciNet  Google Scholar 

  • Zyskind G (1967) On canonical forms, non-negative covariance matrices and best and simple least squares linear estimators in linear models. Ann Math Statist 38:1092–1109

    MathSciNet  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Department of Mathematics, Academy of Economic Studies, Bucharest, Calea Dorobanţi 11-13, Bucharest, Romania

    Gabriela Beganu

Authors
  1. Gabriela Beganu
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Gabriela Beganu.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Beganu, G. Quadratic estimators of covariance components in a multivariate mixed linear model. Stat. Meth. & Appl. 16, 347–356 (2007). https://doi.org/10.1007/s10260-006-0043-3

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10260-006-0043-3

Keywords

Search

Navigation