Influence Diagnostics for a Class of Biased Estimators in Regression

  • Esteban Walker
Conference paper

Abstract

Fractional Principal Components (FPC) is a large class of estimators that include, among others, least squares (LS), ridge, principal components, fractional rank, and Stein estimators. This paper addresses the measurement of influence for FPC estimators. It is shown that the residual and the leverage of any observation depends directly on the “fraction” of the principal components that is used. Generalizations of popular measures of influence for LS are proposed. Approximate deletion formulas are presented.

Keywords

Stein 

References

  1. Belsley, D. A., Kuh, E. and Welsch, R. E. (1980) Regression Diagnostics, Wiley, NY.MATHCrossRefGoogle Scholar
  2. Cook, R. D. and Weisberg, S. (1982) Residuals and Influence in Regression, Chapman and Hall, NY.MATHGoogle Scholar
  3. Hocking, R. R., Speed, F. M. and Lynn, M. J. (1976) A Class of Biased Estimators in Linear Regression, Technometrics, 18, 425–437.MathSciNetMATHCrossRefGoogle Scholar
  4. Lee, W. and Birch, J. B. (1988) Fractional Principal Components Regression: A General Approach to Biased Estimators, Communications in Statistics, Simulation and Computation, 17, 713–727.MathSciNetMATHCrossRefGoogle Scholar
  5. Mason, R. L. and Gunst, R. F. (1985) Outlier-Induced Collinearities, Technometrics, 27, 401–407.CrossRefGoogle Scholar
  6. Walker, E. and Birch, J. B. (1988) Influence Measures in Ridge Regression, Technometrics, 30, 21–27.CrossRefGoogle Scholar
  7. Walker, E. (1990) Influence Diagnostics for Fractional Principal Components Estimators, Communications in Statistics, Simulation and Computation, 19(3), 919–933.MathSciNetMATHCrossRefGoogle Scholar

Copyright information

© Springer-Verlag New York, Inc. 1992

Authors and Affiliations

  • Esteban Walker
    • 1
  1. 1.Dept. of StatisticsU. of TennesseeKnoxvilleUSA

Personalised recommendations