Statistics and Computing

, Volume 2, Issue 3, pp 117–119 | Cite as

Approximating the Shapiro-Wilk W-test for non-normality

  • Patrick Royston
Papers

Abstract

A new approximation for the coefficients required to calculate the Shapiro-WilkW-test is derived. It is easy to calculate and applies for any sample size greater than 3. A normalizing transformation for theW statistic is given, enabling itsP-value to be computed simply. The distribution of the new approximation toW agrees well with published critical points which use exact coefficients.

Keywords

Non-normality Shapiro-WilkW-test 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Davies, C. S. and Stephens, M. A. (1978) Algorithm AS128. Approximating the covariance matrix of normal order statistics.Applied Statistics,27, 206–212.Google Scholar
  2. Royston, J. P. (1982a) An extension of Shapiro and Wilk'sW test for normality to large samples.Applied Statistics,31, 115–124.Google Scholar
  3. Royston, J. P. (1982b) Algorithm AS177. Expected normal order statistics (exact and approximate).Applied Statistics,31, 161–165.Google Scholar
  4. Royston, J. P. (1983) Some techniques for assessing multivariate normality based on the Shapiro-WilkW.Applied Statistics,32, 121–133.Google Scholar
  5. Royston, P. (1991a) Estimation, reference ranges and goodness-of-fit for the three-parameter lognormal distribution.Statistics in Medicine, in press.Google Scholar
  6. Royston, P. (1991b) Aspects of goodness-of-fit tests for normality. I. Use of the tests when the data contain ties.Technometrics, submitted.Google Scholar
  7. Shapiro, S. S. and Francia, R. S. (1972) An approximate analysis of variance test for normality.Journal of the American Statistical Association,67, 215–216.Google Scholar
  8. Shapiro, S. S. and Wilk, M. B. (1965) An analysis of variance test for normality (complete samples).Biometrika,52, 591–611.Google Scholar
  9. Shea, B. L. and Scallan, A. J. (1988) AS R72. A remark on Algorithm AS128: approximating the covariance matrix of normal order statistics.Applied Statistics,37, 151–155.Google Scholar
  10. Verrill, S. and Johnson, A. (1988) Tables and large-sample distribution theory for censored-data correlation statistics for testing normality.Journal of the American Statistical Association,83, 1192–1197.Google Scholar
  11. Weisberg, S. and Bingham, C. (1975) An approximate analysis of variance test for non-normality suitable for machine calculation.Technometrics,17, 133–134.Google Scholar

Copyright information

© Chapman & Hall 1992

Authors and Affiliations

  • Patrick Royston
    • 1
  1. 1.Department of Medical PhysicsRoyal Postgraduate Medical SchoolLondon

Personalised recommendations