DOI QR코드

DOI QR Code

Outliers in Multivariate Box-Cox Transformed Data

  • Published : 2004.04.01

Abstract

The sensitivity of the multivariate Box-Cox transformation model to simultaneous perturbations of all the data based on the likelihood displacement is studied in order to detect outliers. An example is given for illustration.

References

  1. Biometrics v.27 Transformations of multivariate data Andrew,D.F.;Gnanadesikan,R.;Warner,J.L. https://doi.org/10.2307/2528821
  2. Journal of the Royal Statistical Society (B) v.26 An analysis of transformations (with discussions) Box,G.E.P.;Cox,D.R.
  3. Journal of the Royal Statistical Society (B) v.48 Assessment of local influence (with discussions) Cook,R.D.
  4. Applird Multivariate Statistical Analysis(4th ed.) Johnson,R.A.;Wichern,D.W.
  5. Matrix Differential Calculus with Applications in Statistics and Econometrics Magnus,J.R.;Neudecker,H.
  6. Technometrics v.42 Robust diagnostic data analysis: transformations in regression (with discussion) Riani,M.;Atkinson,A.C. https://doi.org/10.2307/1270948
  7. Matrix Analysis for Statistics Schott,J.R.
  8. Journal of the American Statistical Association v.90 Diagnostics and robust estimation in multivariate data transformations Velilla,S. https://doi.org/10.2307/2291329