Skip to main content
SpringerLink
Log in

Invariant tests for multivariate normality: a critical review

  • Survey article
  • Published:
Statistical Papers Aims and scope Submit manuscript

Abstract

This paper gives a synopsis on affine invariant tests of the hypothesis that the unknown distribution of a d-dimensional random vector X is some nondegenerate d-variate normal distribution, on the basis of i.i.d. copies X 1,...,X n of X. Particular emphasis is given to progress that has been achieved during the last decade. Furthermore, we stress the typical diagnostic pitfall connected with purportedly ‘directed’ procedures, such as tests based on measures of multivariate skewness.

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

  • Ahn, S.K. (1992). F-probability plot and its application to multivariate normality. Commun. Statist. — Th. Meth., 21, 997–1023.

    Article  MATH  Google Scholar 

  • Anderson, H.H., Hall, P. and Titterington, D.M. (1994). Two-sample test statistics for measuring discrepancies between two multivariate probability density functions using kernel-based density estimates. Corny. Statist. Data Anal., 30, 119–142.

    Google Scholar 

  • Baringhaus, L., and Henze, N. (1988). A consistent test for multivariate normality based on the empirical characteristic function. Metrika, 35, 339–348.

    Article  MATH  MathSciNet  Google Scholar 

  • Baringhaus, L., and Henze, N. (1991). Limit distributions for measures of multivariate skewness and kurtosis based on projections. J. Multiv. Anal., 38, 51–69.

    Article  MATH  MathSciNet  Google Scholar 

  • Baringhaus, L., and Henze, N. (1992). Limit distributions for Mardia’s measure of multivariate skewness. Ann. Statist., 20, 1889–1902.

    Article  MATH  MathSciNet  Google Scholar 

  • Baxter, M.J., and Gale, N.H. (1998). Testing for multivariate normality via univariate tests: A case study using lead isotope ratio data. J. Appl. Statist., 25, 671–683.

    Article  MATH  Google Scholar 

  • Beirlant, J., Mason, D.M. and Vynckier, C. (1999). Goodness-of-fit analysis for multivariate normality based on generalized quantiles. Comp. Statist. Data Anal., 30, 119–142.

    Article  MATH  MathSciNet  Google Scholar 

  • Bera, A., and John, S. (1983). Tests for multivariate normality with Pearson alternatives. Comm. Statist. — Th. Meth., 12, 103–117.

    Article  Google Scholar 

  • Bogdan, M. (1999). Data driven smooth tests for bivariate normality. J. Multiv. Anal., 68, 26–53.

    Article  MATH  MathSciNet  Google Scholar 

  • Bowman, A.W., and Foster, P.J. (1993). Adaptive smoothing and density-based tests of multivariate normality. J. Amer. Statist. Ass., 88, 529–537.

    Article  MATH  MathSciNet  Google Scholar 

  • Cox, D.R., and Small, N.J.H. (1978). Testing multivariate normality. Biometrika, 65, 263–272.

    Article  MATH  Google Scholar 

  • Cox, D.R., and Wermuth, N. (1994). Tests of linearity, multivariate normality and the adequacy of linear scores. Appl. Statist., 43, 347–355.

    Article  MATH  Google Scholar 

  • Csorgo, S. (1986). Testing for normality in arbitrary dimension. Ann. Statist., 14, 708–723.

    Article  MathSciNet  Google Scholar 

  • Csorgo, S. (1989). Consistency of some tests for multivariate normality. Metrika, 36, 107–116.

    Article  MathSciNet  Google Scholar 

  • D’Agostino, R.B., and Stephens, M.A. (eds) (1986). Goodness-of-fit techniques. Marcel Dekker, New York.

    MATH  Google Scholar 

  • De Wet, T., and Randies, R.H. (1987). On the effect of substituting parameter estimators in limiting χ2, U and V statistics. Ann. Statist., 15, 398–412.

    Article  MATH  MathSciNet  Google Scholar 

  • De Wet, T., Venter, J.H., and Van Wyk, J.W.J. (1979). The null distributions of some test criteria of multivariate normality. S. Afr. Statist. J., 13, 153–177.

    MATH  Google Scholar 

  • Dunn, Ch.L. (1995). Critical values and powers for tests of uniformity of directions under multivariate normality. Comm. Statist. — Th. Meth., 24, 2541–2560.

    Article  MATH  MathSciNet  Google Scholar 

  • Eaton, M.L.(1989). Group invariance applications in statistics. Regional Conf. Ser. in Probab. and Statist. Vol. 1. Inst, of Mathem. Statistics.

  • Eaton, M.L., and Perlman, M.D. (1973). The non-singularity of generalized sample covariance matrices. Ann. Statist., 1, 710–717.

    Article  MATH  MathSciNet  Google Scholar 

  • Epps, T.W. (1999). Limiting behavior of the ICF test for multivariate normality under Gram-Charlier alternatives. Statist & Prob. Lett., 42, 175–184.

    Article  MATH  MathSciNet  Google Scholar 

  • Epps, T.W., and Pulley, L.B. (1983). A test for normality based on the empirical characteristic function. Biometrika, 70, 723–726.

    Article  MATH  MathSciNet  Google Scholar 

  • Fang, K.T., Li, R.Z. and Liang, J.J. (1998). A multivariate version of Ghosh’s T 3-plot to detect non-multinormality. Comp. Statist. Data Anal., 28, 371–386.

    Article  MATH  MathSciNet  Google Scholar 

  • Fang, K.T., and Wang, Y. (1993). Number-theoretic methods in ststistics. Monographs on statistics and applied probability. Chapman and Hall, London.

    Google Scholar 

  • Fattorini, L. (1986). Remarks on the use of the Shapiro-Wilk statistic for testing multivariate normality. Statistica, 46, 209–217.

    MATH  MathSciNet  Google Scholar 

  • Fattorini, L. (1988). On a large class of tests for multivariate normality. Quaderni dell’ Istituto di Statistica Universita’ degli Studi di Siena, Facolta’ di Scienze Economiche e Bancarie, Vol. 66, Siena, Italy.

  • Fattorini, L. (2001). On the assessment of multivariate normality. Atti della XL Riunoine della Societa Italiana die Statistica, Firenze, 26–28 Aprile 2001, 313–324, Italy.

  • Fattorini, L., and Pisani, C. (2000). Assessing multivariate normality on the “worst” sample configuration. Metron, 58, 23–38.

    MATH  MathSciNet  Google Scholar 

  • Geary, R.C. (1947). Testing for normality. Biometrika, 34, 209–242.

    MATH  MathSciNet  Google Scholar 

  • Gnanadesikan, R. (1977). Methods for Statistical Data Analysis of Multivariate Observations. Wiley, New York.

    MATH  Google Scholar 

  • Gnanadesikan, R., and Kettenring, J.R. (1972). Robust estimates, residuals, and outlier detection with multiresponse data. Biometrics, 28, 81–124.

    Article  Google Scholar 

  • Gürtler, N. (2000). Asymptotic theorems for the class of BHEP-tests for multivariate normality with fixed and variable smoothing parameter (in German). Doctoral dissertation, University of Karlsruhe, Germany.

  • Gutjahr, S., Henze, N., and Folkers, M. (1999). Shortcomings of generalized affine invariant skewness measures. J. Mult. Anal., 71, 1–23.

    Article  MATH  MathSciNet  Google Scholar 

  • Hall, P. (1984). Central limit theorem for integrated square error of multivariate nonparametric density estimators. J. Multiv. Anal., 14, 1–15.

    Article  MATH  Google Scholar 

  • Hasofer, A.M., and Stein, G.Z. (1990). Testing for multivariate normality after coordinate transformation. Comm. Statist. — Th. Meth., 19, 1403–1418.

    Article  MathSciNet  Google Scholar 

  • Henze, N. (1990). An approximation to the limit distribution of the Epps-Pulley test statistic for normality. Metrika, 37, 7–18.

    Article  MATH  MathSciNet  Google Scholar 

  • Henze, N. (1994a). On Mardia’s kurtosis test for multivariate normality. Comm. Statist. — Th. Meth., 23, 1031–1045.

    Article  MATH  MathSciNet  Google Scholar 

  • Henze, N. (1994b). The asymptotic behavior of a variant of multivariate kurtosis. Comm. Statist. — Th. Meth., 23, 1047–1061.

    Article  MATH  MathSciNet  Google Scholar 

  • Henze, N. (1994c). Tests for normality (in German). Allg. Statistisches Archiv, 78, 293–317.

    Google Scholar 

  • Henze, N. (1997a). Limit laws for multivariate skewness in the sense of Móri, Rohatgi and Szekely. Statist. & Prob. Lett., 33, 299–307.

    Article  MATH  MathSciNet  Google Scholar 

  • Henze, N. (1997b). Do components of smooth tests of fit have diagnostic properties? Metrika, 45, 121–130.

    Article  MATH  Google Scholar 

  • Henze, N. (1997c). Extreme smoothing and testing for multivariate normality. Statist. & Prob. Lett., 35, 203–213.

    Article  MATH  MathSciNet  Google Scholar 

  • Henze, N., and Klar, B. (1996). Properly rescaled components of smooth tests of fit are diagnostic. Austral. J. Statist., 38, 61–74.

    Article  MATH  MathSciNet  Google Scholar 

  • Henze, N., and Wagner, Th. (1997). A new approach to the BHEP tests for multivariate normality. J. Multiv. Anal., 62(1), 1–23.

    Article  MATH  MathSciNet  Google Scholar 

  • Henze, N., and Zirkler, B. (1990). A class of invariant and consistent tests for multivariate normality. Coram. Statist. — Th. Meth., 19, 3595–3617.

    Article  MATH  MathSciNet  Google Scholar 

  • Horswell, R.L., and Looney, St.W. (1992a). Diagnostic limitations of skewness coefficients in assessing departures from univariate and multivariate normality. Comm. Statist. — Simul., 22, 437–439.

    Article  Google Scholar 

  • Horswell, R.L., and Looney, St.W. (1992b). A comparison of tests for multivariate normality that are based on measures of multivariate skewness and kurtosis. J. Statist. Comp. Simul., 42, 21–38.

    Article  Google Scholar 

  • Isogai, T. (1982). On a measure of multivariate skewness and a test for multivariate normality. Ann. Inst. Statist. Math., 34, 531–541.

    Article  MATH  MathSciNet  Google Scholar 

  • Isogai, T. (1983). On measures of multivariate skewness and kurtosis. Math. Japonica, 28, 251–261.

    MATH  MathSciNet  Google Scholar 

  • Isogai, T. (1989). On using influence functions for testing multivariate normality. Ann. Inst. Statist. Math., 41, 169–186.

    MATH  MathSciNet  Google Scholar 

  • Janssen, A. (2000). Global power functions of goodness of fit tests. Ann. Statist., 28, 239–253.

    Article  MATH  MathSciNet  Google Scholar 

  • Kallenberg, W.C.M., Ledwina, T., and Rafajlowicz, E. (1997). Testing bivariate independence and normality. Sankhya Ser. A, 59, 42–59.

    MATH  MathSciNet  Google Scholar 

  • Kariya, T., and George, E.I. (1994). Locally best invariant tests for multivariate normality in curved families with μ known. IMS Lecture Notes — Monograph Series, 24, 311–322.

    Article  MathSciNet  Google Scholar 

  • Kariya, T., and George, E.I. (1995). Locally best invariant tests for multivariate normality in curved families and Mardias’s test. Sankhya Ser. A, 57, 440–451.

    MATH  MathSciNet  Google Scholar 

  • Kariya, T., Tsai, R.S., Terui, N., and Li, H. (1999). Tests for multinormality with applications to time series. Comm. Statist. — Th. Meth., 28, 519–536.

    Article  MATH  Google Scholar 

  • Klar, B. (1998). Classical and new goodness-of-fit tests (in German). Doctoral dissertation, University of Karlsruhe, Germany.

  • Koziol, J.A. (1982). A class of invariant procedures for assessing multivariate normality. Biometrika, 69, 423–427.

    Article  MATH  MathSciNet  Google Scholar 

  • Koziol, J.A. (1983). On assessing multivariate normality. J. Roy. Statist. Soc. Ser. B, 45, 358–361.

    MATH  MathSciNet  Google Scholar 

  • Koziol, J.A. (1986a). A note on the asymptotic distribution of Mardia’s measure of multivariate kurtosis. Comrn. Statist. — Th. Meth., 15, 1507–1513.

    Article  MATH  MathSciNet  Google Scholar 

  • Koziol, J.A. (1986b). Assessing multivariate normality: A compendium. Comm. Statist. — Th. Meth., 15, 2763–2783.

    Article  MATH  MathSciNet  Google Scholar 

  • Koziol, J.A. (1987). An alternative formulation of Neyman’s smooth goodness of fit tests under composite alternatives. Metrika, 34, 17–24.

    Article  MATH  MathSciNet  Google Scholar 

  • Koziol, J.A. (1989). A note on measures of multivariate kurtosis. Biom. J., 31, 619–624.

    Article  MathSciNet  Google Scholar 

  • Koziol, J.A. (1993). Probability plots for assessing multivariate normality. The Statistician, 42, 161–173.

    Article  Google Scholar 

  • Kuwana, Y., and Kariya, T. (1991). LBI tests for multivariate normality in exponential power distributions. J. Multiv. Anal., 39, 117–134.

    Article  MATH  MathSciNet  Google Scholar 

  • Liang, J., and Bentler, P.M. (1999). A t-distribution plot to detect non-multinormality. Comput. Statist. Data Anal., 30, 31–44.

    Article  MATH  MathSciNet  Google Scholar 

  • Liang, J., Li, R., Fang, H., and Fang, K.T. (2000). Testing multinormality based on lowdimensional projection. J. Statist. Plann. Infer., 86, 129–141.

    Article  MATH  MathSciNet  Google Scholar 

  • Liu, Y., and Cheng, P. (1997). The limit distribution of PP Cramer-von Mises normality test statistics for high dimensions and large sample size. Northeast. Math. J., 13, 213–228.

    MATH  MathSciNet  Google Scholar 

  • Machado, S.G. (1983). Two statistics for testing for multivariate normality. Biometrika, 70, 713–718.

    Article  MATH  MathSciNet  Google Scholar 

  • Malkovich, J.F., and Afifi, A.A. (1973). On tests for multivariate normality. J. Amer. Statist. Ass., 68, 176–179.

    Article  Google Scholar 

  • Mardia, K.V. (1970). Measures of multivariate skewness and kurtosis with applications. Biometrika, 57, 519–530.

    Article  MATH  MathSciNet  Google Scholar 

  • Mardia, K.V. (1974). Applications of some measures of multivariate skewness and kurtosis in testing normality and robustness studies. Sankhya Ser. B, 36, 115–128.

    MATH  MathSciNet  Google Scholar 

  • Mardia, K.V. (1975). Assessment of multinormality and the robustness of Hotelling’s T 2-test. Appl. Statist., 24, 163–171.

    Article  MathSciNet  Google Scholar 

  • Mardia, K.V. (1980). Tests of univariate and multivariate normality. In: Handbook of Statistics 1. Analysis of Variance (P.R. Krishnaiah, ed., North-Holland, Amsterdam., 279–290.

    Google Scholar 

  • Mardia, K.V., and Kanazawa, M. (1983). The null distribution of multivariate kurtosis. Comm. Statist. — Simul., 12, 569–576.

    Article  MATH  MathSciNet  Google Scholar 

  • Mardia, K.V., and Kent, J.T. (1991). Rao score tests for goodness of fit and independence. Biometrika, 78, 355–363.

    Article  MATH  MathSciNet  Google Scholar 

  • Moore, D., and Stubblebine, J. (1981). Chi-square tests for multivariate normality with applications to common stock prices. Comm. Statist. — Th. Meth., 10, 713–738.

    Article  MathSciNet  Google Scholar 

  • Móri, T.F., Rohatgi, V.K., and Székely, G.J. (1993). On multivariate skewness and kurtosis. Th. Prob. Appl., 38, 547–551.

    Article  Google Scholar 

  • Mudholkar, G.S., McDermott, M., and Srivastava, D.K. (1992). A test of p-variate normality. Biometrika, 79, 850–854.

    Article  MATH  MathSciNet  Google Scholar 

  • Mudholkar, G.S., Srivastava, D.K., and Lin, C.Th. (1995). Some p-variate adaptions of the Shapiro-Wilk test of normality. Comm. Statist. — Th. Meth., 24, 953–985.

    Article  MATH  MathSciNet  Google Scholar 

  • Naito, K. (1996). On weighting the studentized empirical characteristic function for testing normality. Comm. Statist. — Simul., 25, 201–213.

    Article  MATH  MathSciNet  Google Scholar 

  • Naito, K. (1998). Approximation of the power of kurtosis test for multinormality. J. Multiv. Anal., 65, 201–213.

    Article  MathSciNet  Google Scholar 

  • Oja, H. (1983). Descriptive statistics for multivariate distributions. Statist. Prob. Lett., 1, 327–332.

    Article  MATH  MathSciNet  Google Scholar 

  • Ozturk, A., and Romeu, J.L. (1992). A new method for assessing multivariate normality with graphical applications. Comm. Statist. — Simula., 21, 15–34.

    Article  MATH  Google Scholar 

  • Park, C. (1999). A note on the chi-square test for multivariate normality based on the sample Mahalanobis distances. J. Korean Statist. Soc., 28, 479–488.

    MathSciNet  Google Scholar 

  • Paulson, A.S., Roohan, P., and Sullo, P. (1987). Some empirical distribution function tests for multivariate normality. J. Statist. Comput. Simul., 28, 15–30.

    Article  MATH  Google Scholar 

  • Prentice, M.J. (1978). On invariant tests of uniformity for directions and orientations. Ann. Statist., 6, 169–176.

    Article  MATH  MathSciNet  Google Scholar 

  • Quiroz, A.J., and Dudley, R.M. (1991). Some new tests for multivariate normality. Prob. Th. Rel. Fields, 87, 521–546.

    Article  MATH  MathSciNet  Google Scholar 

  • Quiroz, A.J., and Trabucco, J.C. (1998). Kolmogorov-Smirnov type statistics for testing multivariate normality. Publ. Mat. Urug., 7, 67–82.

    MathSciNet  Google Scholar 

  • Rao, C.R., and Ali, H. (1998). An overall test for multivariate normality. Student, 2, 317–324.

    Google Scholar 

  • Rayleigh, L (1919). On the problems of random vibrations and of random flights in one, two or three dimensions. Phil. Mag., 57, 321–347.

    Google Scholar 

  • Rayner, J.C.W., and Best, D.J. (1989). Smooth tests of goodness of fit. Oxford University Press, New York.

    MATH  Google Scholar 

  • Rayner, J.C.W., Best, D.J., and Matthews, K.L. (1995). Interpreting the skewness coefficient. Coram. Statist. — Th. Meth., 24, 593–600.

    Article  MATH  Google Scholar 

  • Romeu, J.L. (1992). A comparative visualization of a graphical multivariate normality GOF test. Computing Science and Statistics. Proc. 24rd Symp. Interface, 305–309.

  • Romeu, J.L., and Ozturk, A. (1993). A comparative study of goodness-of-fit tests for multivariate normality. J. Multiv. Anal., 46, 309–334.

    Article  MATH  MathSciNet  Google Scholar 

  • Romeu, J.L., and Ozturk, A. (1996). A new graphical test for multivariate normality. Amer. J. Manage. Sci., 16, 5–48.

    MATH  MathSciNet  Google Scholar 

  • Roy, S.N. (1953). On a heuristic method of test construction and its use in multivariate analysis. Ann. Math. Statist., 24, 220–238.

    Article  MATH  Google Scholar 

  • Ruymgaart, F. (1998). A note on weak convergence of density estimators in Hilbert spaces. Statistics, 30, 331–343.

    Article  MATH  MathSciNet  Google Scholar 

  • Schwager, S.J., and Margolin, B.H. (1982). Detection of multivariate normal outliers. Ann. Statist., 10, 943–954.

    Article  MATH  MathSciNet  Google Scholar 

  • Singh, A. (1993). Omnibus robust procedures for assessment of multivariate normality and detection of multivariate outliers. In: Multivariate Environmental Statistics (G.P. Patil and C.R. Rao, eds.) North-Holland, Amsterdam, 445–488.

    Google Scholar 

  • Srivastava, M.S. (1984). A measure of skewness and kurtosis and a graphical method for assessing multivariate normality. Statist. Prob. Lett., 2, 263–267.

    Article  Google Scholar 

  • Sun, J. (1993). Tail probabilities of the maxima of Gaussian random fields. Ann. Prob., 21, 34–71.

    Article  MATH  Google Scholar 

  • Szkutnik, Z. (1988). Most powerful invariant tests for binormality. Ann. Statist., 16, 292–301.

    Article  MATH  MathSciNet  Google Scholar 

  • Tsai, K.T., and Koziol, J.A (1988). A correlation type procedure for assessing multivariate normality. Comm. Statist. — Simula., 17, 637–651.

    Article  MATH  MathSciNet  Google Scholar 

  • Vasicek, O. (1976). A test for normality based on sample entropy. J. Roy. Statist. Soc. A, 139, 54–59.

    MathSciNet  Google Scholar 

  • Versluis, C. (1996). Comparison of tests for bivariate normality with unknown parameters by transformation to a univariate statistic. Comm. Statist. — Th. Meth., 25, 647–665.

    Article  MATH  MathSciNet  Google Scholar 

  • Viollaz, A.J. (1995). Convergence in weigthed supremum norms of the Skorokhod representation of the estimated empirical process. Comm. Statist. — Th. Meth., 24, 2829–2839.

    Article  MATH  MathSciNet  Google Scholar 

  • Wagner, Th. (1995). Asymptotic results on a class of tests for multivariate normality (in German). Diploma thesis, University of Karlsruhe, Germany.

  • Zhu, L., Fang, K.T., and Bhatti, M.I. (1997). On estimated projection pursuit Cramer-von Mises statistics. J. Multiv. Anal., 63, 1–14.

    Article  MATH  MathSciNet  Google Scholar 

  • Zhu, L., Wong, H.L., and Fang, K.T. (1995). A test for multivariate normality based on sample entropy and projection pursuit. J. Statist. Plann. Infer., 45, 373–385.

    Article  MATH  MathSciNet  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Institut für Mathematische Stochastik, Universität Karlsruhe, Englerstraβe 2, 76128, Karlsruhe, Germany

    Norbert Henze

Authors
  1. Norbert Henze
    View author publications

    You can also search for this author in PubMed Google Scholar

Rights and permissions

Reprints and permissions

About this article

Cite this article

Henze, N. Invariant tests for multivariate normality: a critical review. Statistical Papers 43, 467–506 (2002). https://doi.org/10.1007/s00362-002-0119-6

Download citation

  • Received:

  • Revised:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00362-002-0119-6

Keywords and Phrases

AMS Subject Classifications

Search

Navigation