An effective selection of regression variables when the error distribution is incorrectly specified

  • Wolfgang Härdle


An asymptotically efficient selection of regression variables is considered in the situation where the statistician estimates regression parameters by the maximum likelihood method but fails to choose a likelihood function matching the true error distribution. The proposed procedure is useful when a robust regression technique is applied but the data in fact do not require that treatment. Examples and a Monte Carlo study are presented and relationships to other selectors such as Mallows'Cp are investigated.

Key words and phrases

Variable selection regression analysis robust regression model choice 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    Akaike, H. (1970). Statistical predictor identification,Ann. Inst. Statist. Math.,22, 203–217.MathSciNetCrossRefGoogle Scholar
  2. [2]
    Andrews, D. F., Bickel, P. J., Hampel, F., Huber, P., Rogers, W. and Tukey, J. W. (1972).Robust Estimation of Location, Princeton University Press, Princeton.zbMATHGoogle Scholar
  3. [3]
    Breiman, L. and Freedman D. (1983). How many variables should be entered in a regression equation.J. Amer. Statist. Ass.,78, 131–136.MathSciNetCrossRefGoogle Scholar
  4. [4]
    Cox, D. (1983). Asymptotics forM-type smoothing splines,Ann. Statist.,11, 530–551.MathSciNetCrossRefGoogle Scholar
  5. [5]
    Härdle, W. (1984). Robust regression function estimation,J. Multivariate Anal.,14, 169–180.MathSciNetCrossRefGoogle Scholar
  6. [6]
    Huber, P. (1973). Robust regression: Asymptotics, conjectures and Monte Carlo,Ann. Statist.,1, 799–821.MathSciNetCrossRefGoogle Scholar
  7. [7]
    Huber, P. (1981).Robust Statistics, Wiley, New York.CrossRefGoogle Scholar
  8. [8]
    Li, K. C. (1984). Asymptotic optimality forCp,C1, cross-validation and generalized cross-validation: Discrete, index set, Manuscript.Google Scholar
  9. [9]
    Mallows, C. (1973). Some comments onCp,Technometrics,15, 661–675.zbMATHGoogle Scholar
  10. [10]
    Ronchetti, E. (1985). Robust model selection in regression,Statist. Prob. Letters,3, 21–23.MathSciNetCrossRefGoogle Scholar
  11. [11]
    Schrader, R. M. and Hettmansperger, T. P. (1980). Robust analysis of variance based upon a likelihood ratio criterion,Biometrika,67, 93–101.MathSciNetCrossRefGoogle Scholar
  12. [12]
    Shibata, R. (1981). An optimal selection of regression variables,Biometrika,68, 45–54.MathSciNetCrossRefGoogle Scholar
  13. [13]
    Stone, C. J. (1984). An asymptotically optimal window selection rule for kernel density estimates,Ann. Statist.,12, 1285–1297.MathSciNetCrossRefGoogle Scholar
  14. [14]
    Whittle, P. (1960). Bounds for the moments of linear and quadratic forms in independent variables.Theor. Prob. Appl.,3, 302–305.CrossRefGoogle Scholar

Copyright information

© The Institute of Statistical Mathematics, Tokyo 1987

Authors and Affiliations

  • Wolfgang Härdle
    • 1
    • 2
  1. 1.Johann-Wolfgang-Goethe-UniversitätWest Germany
  2. 2.University of North CarolinaUSA

Personalised recommendations