Abstract
The test statistics of assessing multivariate normality based on Roy’s union-intersection principle (Roy, Some Aspects of Multivariate Analysis, Wiley, New York, 1953) are generalizations of univariate normality, and are formed as the optimal value of a nonlinear multivariate function. Due to the difficulty of solving multivariate optimization problems, researchers have proposed various approximations. However, this paper shows that the (nearly) global solution contrarily results in unsatisfactory power performance in Monte Carlo simulations. Thus, instead of searching for a true optimal solution, this study proposes a functional statistic constructed by the q% quantile of the objective function values. A comparative Monte Carlo analysis shows that the proposed method is superior to two highly recommended tests when detecting widely-selected alternatives that characterize the various properties of multivariate normality.
Similar content being viewed by others
References
Azzalini, A., Capitanio, A.: Statistical applications of the multivariate skew-normal distribution. JRSSB 61, 579–602 (1999)
Fang, K.T., Wang, Y.: Number-Theoretic Methods in Statistics. Chapman & Hall, London (1994)
Farrell, P.J., Salibian-Barrera, M., Naczk, K.: On test of multivariate normality and associated simulation studies. J. Stat. Comput. Simul. 77, 1065–1080 (2007)
Fattorini, L.: Remarks on the use of the Shapiro-Wilk statistic for testing multivariate normality. Statistica 2, 209–217 (1986)
Henze, N., Zirler, B.: A class of invariant consistent tests for multivariate normality. Commun. Stat. Theory Methods 19, 3595–3617 (1990)
Johnson, M.E.: Multivariate Statistical Simulation. Wiley, New York (1987)
Lagarias, J.C., Reeds, J.A., Wright, M.H., Wright, P.E.: Convergence properties of the Nelder-Mead simplex method in low dimensions. SIAM. J. Optim. 9(1), 112–147 (1998)
Malkovich, J.F., Afifi, A.A.: On tests for multivariate normality. J. Am. Stat. Assoc. 68, 176–179 (1973)
Mecklin, C., Mundfrom, D.: A Monte Carlo comparison of the Type I and Type II error rates of tests of multivariate normality. J. Stat. Comput. Simul. 75, 93–107 (2005)
Roy, S.N.: Some Aspects of Multivariate Analysis. Wiley, New York (1953)
Royston, J.P.: Some techniques for assessing multivariate normality based on the Shapiro-Wilk W. Appl. Stat. 32, 121–133 (1983)
Royston, J.P.: Approximating the Shapiro-Wilk W test for non-normality. Stat. Comput. 2, 117–119 (1992)
Shapiro, S.S., Wilk, M.B.: An analysis of variance test for normality (complete samples). Biometrika 52, 591–611 (1965)
Szekely, G.J., Rizzo, M.L.: A new test for multivariate normality. J. Multivar. Anal. 93, 58–80 (2005)
Zhu, L.X., Wong, H.L., Fang, K.T.: A test for multivariate normality based on sample entropy and projection pursuit. J. Stat. Planning Inference 45, 373–385 (1995)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Wang, CC., Hwang, YT. A new functional statistic for multivariate normality. Stat Comput 21, 501–509 (2011). https://doi.org/10.1007/s11222-010-9186-z
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11222-010-9186-z