Statistics and Computing

, Volume 15, Issue 4, pp 281–287 | Cite as

Approximation of power in multivariate analysis

  • Ronald W. ButlerEmail author
  • Andrew T. A. Wood


We consider the calculation of power functions in classical multivariate analysis. In this context, power can be expressed in terms of tail probabilities of certain noncentral distributions. The necessary noncentral distribution theory was developed between the 1940s and 1970s by a number of authors. However, tractable methods for calculating the relevant probabilities have been lacking. In this paper we present simple yet extremely accurate saddlepoint approximations to power functions associated with the following classical test statistics: the likelihood ratio statistic for testing the general linear hypothesis in MANOVA; the likelihood ratio statistic for testing block independence; and Bartlett's modified likelihood ratio statistic for testing equality of covariance matrices.


block independence covariance equality general linear hypothesis hypergeometric functions Laplace approximation power function saddlepoint approximation 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Anderson T.W. 1984. An introduction to multivariate analysis. Second Edition. Wiley, New York.Google Scholar
  2. Booth J.G., Butler R.W., Huzurbazar S., and Wood A.T.A. 1995. Saddlepoint approximations for p-values of some tests of covariance matrices. J. Statist. Comput. Simul. 53: 165–180.Google Scholar
  3. Box G.E.P. 1949. A general distribution theory for a class of likelihood ratio criteria. Biometrika 36: 317–346.PubMedGoogle Scholar
  4. Butler R.W., Booth J.G., and Huzurbazar S. 1993. Saddlepoint approximations for tests of block independence, sphericity, and equal variances and covariances. J.R. Statist. Soc. B 55: 171–184.Google Scholar
  5. Butler R.W., Huzurbazar S., and Booth, J.G. 1992. Saddlepoint approximations for the generalized variance and Wilks' statistic. Biometrika 79: 157–169.Google Scholar
  6. Butler R.W. and Wood A.T.A. 2002. Laplace approximations for hypergeometric functions of matrix argument. Ann. Statist 30: 1155–1177.CrossRefGoogle Scholar
  7. Butler R.W. and Wood A.T.A. 2004. A dimensional CLT for Wilks' lambda in multivariate analysis. Scandinavian Journal of Statistics 31: 585–601.CrossRefGoogle Scholar
  8. Constantine A.G. 1963. Some noncentral distribution problems in multivariate analysis. Ann. Math. Statist. 34: 1270–1285.Google Scholar
  9. Herz C.S. 1955. Bessel functions of matrix argument. Ann. Math. 61: 474–523.Google Scholar
  10. Jensen J.L. 1991. A large deviation type approximation for the ‘Box class’ of likelihood ratio criteria. J. Amer. Statist. Assoc. 86: 437–440.MathSciNetGoogle Scholar
  11. Jensen J.L. 1995. Saddlepoint Approximations. Clarendon Press, Oxford.Google Scholar
  12. James A.T. 1960. The distribution of the latent roots of the covariance matrix. Ann. Math. Statist. 31: 151–158.Google Scholar
  13. James A.T. 1961. Zonal polynomials of the real positive definite symmetric matrices. Ann. Math. 74: 456–469.Google Scholar
  14. Kulp R.W. and Nagarsenkar B.N. 1984. An asymptotic expansion of the nonnull distribution of Wilks criterion for testing the multivariate linear hypothesis Ann. Statist. 12: 1576–1583.Google Scholar
  15. Lee Y.-S. 1971. Distribution of the canonical correlations and asymptotic expansions for distributions of certain independence test statistics. Ann. Math. Statist. 42: 526–537.Google Scholar
  16. Lugananni R. and Rice, S. 1980. Saddlepoint approximations for the distribution of the sum of independent random variables. Adv. Appl. Probab 12: 475–490.Google Scholar
  17. Muirhead R.J. 1972. On the test of independence between two sets of variates. Ann. Math. Statist. 43: 1491–1497.Google Scholar
  18. Muirhead R.J. 1982. Aspects of Multivariate Statistical Theory. John Wiley, New York.Google Scholar
  19. Muller K.E., LaVange L.M., Ramey S.L., and Ramey C.T. 1992. Power calculations for general linear multivariate models including repeated measures applications. J. Amer. Statist. Assoc. 87: 1209–1226.Google Scholar
  20. Muller K.E. and Peterson B.L. 1984. Practical methods for computing power in testing the multivariate general linear hypothesis. Computational Statistics and Data Analysis 2: 143–158.CrossRefGoogle Scholar
  21. Srivastava M.S. and Yao W.K. 1989. Saddlepoint method for obtaining tail probability of Wilks' likelihood ratio test. J. Mult. Anal. 31: 117–126.CrossRefGoogle Scholar
  22. Sugiura N. 1969. Asymptotic non-null distribution of the likelihood ratio criteria for covariance matrix under local alternatives. Mimeo Ser. 609, Institute of Statistics, University of North Carolina, Chapel Hill.Google Scholar
  23. Sugiura N. 1973a. Further asymptotic formulas for the non-null distributions of three statistics for multivariate linear hypothesis. Ann. Inst. Statist. Math. 25: 153–163.Google Scholar
  24. Sugiura N. 1973b. Asymptotic non-null distributions of the likelihood ratio criteria for covariance matrix under local alternatives. Ann. Statist. 1: 718–728.Google Scholar
  25. Sugiura N. 1974. Asymptotic formulas for the hypergeometric function 2 F1 of matrix argument, useful in multivariate analysis. Ann. Inst. Statist. Math. 26: 117–125.Google Scholar
  26. Sugiura N. and Fujikoshi Y. 1969. Asymptotic expansion of the non-null distributions of the likelihood ratio criteria for multivariate linear hypothesis and independence. Ann. Math. Statist. 40: 942–952.Google Scholar

Copyright information

© Springer Science + Business Media, Inc. 2005

Authors and Affiliations

  1. 1.Department of StatisticsColorado State UniversityFort CollinsUSA

Personalised recommendations