Journal of the Korean Statistical Society

The Korean Statistical Society (한국통계학회)
권호 표지

Recent Developments in Discriminant Analysis fro man Information Geometric Point of View

  • Eguchi, Shinto (Institute of Statistical Mathematics) ;
  • Copas, John B. (Department of Statistics, University of Warwick)
  • Published : 2001.06.01
Download PDF
⟨ Previous Next ⟩

Abstract

This paper concerns a problem of classification based on training dta. A framework of information geometry is given to elucidate the characteristics of discriminant functions including logistic discrimination and AdaBoost. We discuss a class of loss functions from a unified viewpoint.

Keywords

References

  1. Lecture notes in statistics v.28 Differential-geometrical methods in statistics Amari, S.
  2. Methods of Information Geometry Translations of Mathematical Monographs v.191 Amari, S.;Nagaoka, H.
  3. Classification and Regression Trees Breiman, L.;Friedman, J.;Olshen, R.;Stone, C.
  4. Biometrics v.23 A general maximum likelihood discriminant Day, N.E.;Kerridge, D.F.
  5. J. Amer. Statist. Assoc. v.70 The efficiency of logistic regression compared to normal discriminant analysis is Efron, B.
  6. Ann. Statist. v.3 Defining the curvature of a statistical problem (with applications to second order efficiency) Efron, B.
  7. revision in Biometrika A class of logistic-type discriminant functions Eguchi, S.;Copas, J.B.
  8. Phi. Trans. R. Soc., A v.222 On the mathematical foundations of the theoretical statistics Fisher, R.A.
  9. Information and Computation v.121 no.2 Boosting a weak learning algorithm by majority Freund, Y.
  10. Ann. Statist. v.28 Additive logistic regression: a statistical view of boosting (with discussion) Friedman, J.;Hastie, T.;Tibshirani, R.
  11. Ann. Statist. v.13 Projection pursuit (with discussion) Huber, P.J.
  12. Ann. Statist. v.23 An infinite-dimensional geometric structure on the space of all the probability measures equivalent to a given one Pistone, G.;Sempi, C.
  13. Machine Learning v.5 The strength of weak learnability Schapire, R.E.
  14. Econometrica v.50 Maximum likelihood estimation of misspecified models White, H.