Advertisement

Statistical Papers

, Volume 44, Issue 1, pp 23–45 | Cite as

On influence diagnostic in univariate elliptical linear regression models

  • Manuel GaleaEmail author
  • Gilberto A. Paula
  • Miguel Uribe-Opazo
Articles

Abstract

We discuss in this paper the assessment of local influence in univariate elliptical linear regression models. This class includes all symmetric continuous distributions, such as normal, Student-t, Pearson VII, exponential power and logistic, among others. We derive the appropriate matrices for assessing the local influence on the parameter estimates and on predictions by considering as influence measures the likelihood displacement and a distance based on the Pearson residual. Two examples with real data are given for illustration.

Key words

Elliptical distributions Likelihood displacement Local influence Pearson residual Robustness 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Andrews, D. F.(1974). A robust method for multiple linear regression. Technometrics 16, 523–531.zbMATHCrossRefMathSciNetGoogle Scholar
  2. Atkinson, A. C. (1985). Plots, Transformations and Regression. Claxendon Press: Oxford.zbMATHGoogle Scholar
  3. Brownlee, K. A. (1965). Statistical Theory and Methodology in Science and Engineering, 2nd Edition. John Wiley & Sons: New York.zbMATHGoogle Scholar
  4. Carroll, R. J. and Ruppert, D. (1985). Transformations in regression: a robust analysis. Technometrics 27, 1–12.zbMATHCrossRefMathSciNetGoogle Scholar
  5. Cook, R. D. (1986). Assessment of local influence (with discussion). J. R. Statist. Soc. B 48, 133–169.Google Scholar
  6. Davison, A.C. and Tsai, C-L. (1992). Regression model diagnostics. Int. Statist. Rev. 60, 337–353.zbMATHCrossRefGoogle Scholar
  7. Fang, K. T. and Anderson, T. W. (1990). Statistical Inference in Elliptical Contoured and Related Distributions. Allerton Press: New York.Google Scholar
  8. Fung, W. K. (1993). Unmasking outliers and leverage points: A Confirmation. J. Amer. Statist. Assoc. 88, 515–519.CrossRefMathSciNetGoogle Scholar
  9. Fung, W. K. and Kwan, C. W. (1997). A note on local influence based on normal curvature. J. R. Statist. Soc. B 59, 839–843.CrossRefMathSciNetGoogle Scholar
  10. Galea, M.; Paula, G. A. and Bolfarine, H. (1997). Local influence in elliptical linear regression models. The Statistician 46, 71–79.Google Scholar
  11. Gu, H. and Fung, W. K. (1998). Assessing local influence in canonical correlation analysis. Ann. Inst. Statist. Math. 50, 755–772.zbMATHCrossRefMathSciNetGoogle Scholar
  12. Kowalski, J.; Mendonza-Blanco, J. R.; Tu, X. M. and Gleser, L. J. (1999). On the difference in inference and prediction between the joint and independent t-error models for seemingly unrelated regressions. Commun. Statist., Theory Meth. 28, 2119–2140.zbMATHCrossRefGoogle Scholar
  13. Kwan, C. W. and Fung, W. K. (1998). Assessing local influence for specific restricted likelihood: Applications to factor analysis. Psychometrika 63, 35–46.zbMATHCrossRefMathSciNetGoogle Scholar
  14. Lange, K. L.; Little, R. J. A. and Taylor, J. M. G. (1989). Robust statistical modeling using the T distribution. J. Amer. Statist. Assoc. 84, 881–896.CrossRefMathSciNetGoogle Scholar
  15. Lesaffre, E. and Verbeke, G. (1998). Local influence in linear mixed models. Biometrics 38, 963–974.Google Scholar
  16. Liu, S. Z. (2000). On local influence for elliptical linear models. Statist. Papers 41, 211–224.zbMATHCrossRefMathSciNetGoogle Scholar
  17. Pan, J. X.; Fang, K. T. and von Rosen (1997). Local influence assessment in the growth curve model with unstructured covariance. J. Statist. Plann. Infer. 62, 263–278.zbMATHCrossRefGoogle Scholar
  18. Paula, G. A. (1995). Influence and residuals in restricted generalized linear models. J. Statist. Comp. Simul. 51, 315–352.CrossRefMathSciNetGoogle Scholar
  19. Rousseeuw, P. J. and van Zomeren, B. C. (1990). Unmasking multivariate outliers and leverage points (with discussion). J. Amer. Statist. Assoc. 85, 633–651.CrossRefGoogle Scholar
  20. Ruppert, D. and Carroll, R. J. (1980). Trimmed least squares estimation in the linear model. J. Amer. Statist. Assoc. 75, 828–838.zbMATHCrossRefMathSciNetGoogle Scholar
  21. Taylor, J. M. G. (1991). Properties of modelling the error distribution with an extra shape parameter. Comput. Statist. Data Anal. 13, 33–46.CrossRefGoogle Scholar
  22. Thomas, W. and Cook, R. D. (1990). Assessing influence on predictions from generalized linear models. Technometrics 32, 59–65.CrossRefMathSciNetGoogle Scholar

Copyright information

© Springer-Verlag 2003

Authors and Affiliations

  • Manuel Galea
    • 1
    Email author
  • Gilberto A. Paula
    • 2
  • Miguel Uribe-Opazo
    • 3
  1. 1.Departamento de EstadísticaUniversidad de ValparaísoValparaísoChile
  2. 2.Instituto de Mathemática e EstatisticaUniversidada de São PauloSão Paulo — SPBrazil
  3. 3.Centra de Ciências Exatas e TecnológicasUniversidade Estadual do Oeste do Paraná, Rua Universitária 119 -Jardim UniversitárioCascavel — PRBrazil

Personalised recommendations