Gini-PLS Regressions

Original Article
  • 5 Downloads

Abstract

Data contamination and excessive correlations between regressors (multicollinearity) constitute a standard and major problem in econometrics. Two techniques enable solving these problems, in separate ways: the Gini regression for the former, and the PLS (partial least squares) regression for the latter. Gini-PLS regressions are proposed in order to treat extreme values and multicollinearity simultaneously.

Keywords

Gini covariance Gini regression Gini-PLS regressions PLS regression 

JEL Classification

C3 C8 

References

  1. Bastien, P., V. Esposito Vinzi, and M. Tenenhaus. 2005. PLS generalised linear regression. Computational Statistics and Data Analysis 48: 17–46Google Scholar
  2. Bry, X., C. Trottier, T. Verron, and F. Mortier. 2013. Supervised component generalized linear regression using a PLS-extension of the Fisher scoring algorithm. Journal of Multivariate Analysis 119: 47–60.CrossRefGoogle Scholar
  3. Choi, S.W. 2009. The effect of outliers on regression analysis: Regime type and foreign direct investment. Quarterly Journal of Political Science 4: 153–165.CrossRefGoogle Scholar
  4. Chung, D., and S. Keles. 2010. Sparse partial least squares classification for high dimensional data. Statistical Applications in Genetics and Molecular Biology 91, article 17.Google Scholar
  5. Dixon, W.J. 1950. Analysis of extreme values. The Annals of Mathematical Statistics 2 (4): 488–506.CrossRefGoogle Scholar
  6. Durbin, J. 1954. Errors in variables. Review of the International Statistical Institute 22: 23–32.CrossRefGoogle Scholar
  7. John, G.H. 1995. Robust decision tree: Removing outliers from databases. In KDD-95 Proceeding, 174–179.Google Scholar
  8. Olkin, I., and S. Yitzhaki. 1992. Gini regression analysis. International Statistical Review 602: 185–196.CrossRefGoogle Scholar
  9. Planchon, V. 2005. Traitement des valeurs aberrantes: concepts actuels et tendances générales. Biotechnologie Agronomie Société et Environnement 91: 185–196.Google Scholar
  10. Russolillo, G. 2012. Non-Metric partial least squares. Electronic Journal of Statistics 6: 1641–1669.CrossRefGoogle Scholar
  11. Tenenhaus, M. 1998. La régression PLS théorie et pratique. Paris: Technip.Google Scholar
  12. Schechtman, E., and S. Yitzhaki. 1999. On the proper bounds of the Gini correlation. Economics Letters 63 (2): 133–138.CrossRefGoogle Scholar
  13. Schechtman, E., and S. Yitzhaki. 2003. A family of correlation coefficients based on extended Gini. Journal of Economic Inequality 1: 129–146.CrossRefGoogle Scholar
  14. Wold, S., C. Albano, W. J. Dunn III, K. Esbensen, S. Hellberg, E. Johansson, and H. Sjöström. 1983. Pattern recognition: Finding and using regularities in multivariate data. In Proc. UFOST Conf., Food Research and Data Analysis, ed. J. Martens. Applied Science Publications: London.Google Scholar
  15. Wold, S., H. Martens, and H. Wold. 1983. The multivariate calibration problem in chemistry solved by the PLS method. In Proc. Conf. Matrix Pencils, eds. A. Rnhe, and B. Kagstroem, 286–293. Berlin: Springer.Google Scholar
  16. Yitzhaki, S. 2003. Gini’s mean difference: A superior measure of variability for non-normal distributions. Metron 61 (2): 285–316.Google Scholar
  17. Yitzhaki, S., and E. Schechtman. 2004. The Gini instrumental variable, or the ’double instrumental variable’ estimator. Metron 52 (3): 287–313.Google Scholar
  18. Yitzhaki, S., and E. Schechtman. 2013. The Gini Methodology: A Primer on a Statistical Methodology. Berlin: Springer.CrossRefGoogle Scholar

Copyright information

© The Indian Econometric Society 2018

Authors and Affiliations

  • Stéphane Mussard
    • 1
    • 2
    • 3
    • 4
  • Fattouma Souissi-Benrejab
    • 5
    • 6
  1. 1.Chrome Université de NîmesNîmesFrance
  2. 2.MRE University of MontpellierMontpellierFrance
  3. 3.GrEdi University of SherbrookeQuebecCanada
  4. 4.Liser LuxembourgEsch-sur-AlzetteLuxembourg
  5. 5.Université Montpellier 1, UMR5474 LAMETAMontpellierFrance
  6. 6.Faculté d’EconomieMontpellier CedexFrance

Personalised recommendations