Journal of the Korean Statistical Society

The Korean Statistical Society (한국통계학회)
권호 표지

Selecting a Transformation to Reduce Skewness

  • Yeo, In-Kwon (Department of Control and Instrumentation Engineering, Kangwon National University)
  • Published : 2001.12.01
Download PDF
⟨ Previous Next ⟩

Abstract

In this paper, we study selecting a transformation so that the transformed variable is nearly symmetrically distributed. The large sample properties of an M-estimator of transformation parameter that is obtained by minimizing the integrated square of the imaginary part of the empirical characteristic function are investigated when a random sample is selected from some unspecified distribution. According to influence function calculations and Monte Carlo simulations, these estimates are less sensitive, than the normal model maximum likelihood estimates, to a few outliers.

Keywords

References

  1. Biometrika v.58 A nore on the selection of data transformations Andrew;D.F.
  2. Journal of the Royal Statistical Society v.B 26 An analysis of transformations Box;G. E. P.;Cox;D. R.
  3. Journal of the Royal Statistical Society v.B 42 A robust method for testing transformations to achieve approximate normality Carroll;R. J.
  4. Journal of the American Statistical Association v.75 The large-sample behavior of trnasformations to normality Hernadez;F.;Johnson;R. A.
  5. Biometrika v.62 On power transformations to symmetry Hinkley;D. V.
  6. U-statistics : Theory and Practice Lee;A. J.
  7. Annals of Mathematical Statistics v.27 Uniform convergence of random functions with applications to statistics Rubin;H.
  8. Biometrika v.72 Power Transformations to Symmetry Taylor;J. M. G.
  9. Convex transformations of random variables van Zwet;W. R.
  10. Biometrika v.87 A new family of power transformations to improve normality or symmetry Yeo;I. K.;Johnson;R. A.
  11. Statitics and Probabilty Letters v.51 A uniform strong law of large numbers for U-statistics with application to transforming to near symmetry Yeo;I.K.;Johnson;R. A.