On an Optimal Bayesian Variable Selection Method for Generalized Logit Model

  • Published : 2000.08.01

Abstract

This paper is concerned with suggesting a Bayesian method for variable selection in generalized logit model. It is based on Laplace-Metropolis algorithm intended to propose a simple method for estimating the marginal likelihood of the model. The algorithm then leads to a criterion for the selection of variables. The criterion is to find a subset of variables that maximizes the marginal likelihood of the model and it is seen to be a Bayes rule in a sense that it minimizes the risk of the variable selection under 0-1 loss function. Based upon two examples, the suggested method is illustrated and compared with existing frequentist methods.

References

  1. Biometrika v.71 Calculation of polychotomous logistic regression parameters using individualized regression Begg, C.B;Gray, R.
  2. Journal of American Statistics Association v.91 The Intrinsic Bayes Factor for Model Selection and Prediction Berger, J.O.;Pericchi, L.R.
  3. Using the SAS System Logistic Regression Example Jack J. Berry;Duane S. Hayes;Kristin R. Latour;David C. Schlotzhauer
  4. Journal of American Statistics Association v.90 Marginal Likelihood from the Gibbs Output Chib, S.
  5. Journal of American Statistics v.49 Understanding the Metropolis-Hasting Algorithm Chib, S.;Greenberg, E.
  6. A Guide to The Analysis of Categorical Depenent variables with SAS and SPSS Choong Geun Chung
  7. Journal of American Statistics Association v.91 Markov Chain Mote Carlo Convergence Diagnostics : A Comparitive Review Cowles, M.K.;Carlin, B.P.
  8. Applied Logistic Regression David W. Hosmer, J.
  9. Asymptotic Methods in Analysis de Bruijn, N.G.
  10. Statistical Science v.7 Inference from iterative simulation using multiple sequence(with disscessed) Gelman, A.E.;Rubin, D.B.
  11. Estimating Bayes Factors via Posterior Simulation With the Laplace-Metropolis Estimator v.82 no.438 Lewis, S.M.;Raftery, A.E.
  12. A Genertic Approach to Posterior Intergration Gibbs Sampling Muller, P.
  13. Technical Report, University of Cambridge Weak Convergence and Optimal Scaling of Random Walk Metropolis Algorithms Roberts, G.O.;Gelman, A.;Gilks, W.R.
  14. Journal of American Statistics Association v.85 Unmasking multivariate outliers and leverage points(with discussion) Rousseeuw, P.J.;van Zomeren, B.C.
  15. Instruduction Statistical Review v.58 A survey of multivariate medians Small, G.G.
  16. Annals of Statistics v.22 Markov Chains for Exploring Posterior Distributions Tierney, L.
  17. Journal of American Statistics Association v.81 Accurate Approximations for Posterior Moments and Marginal Densities Tierney, L.;Kadane, J.B.
  18. Medical Decision Masking v.3 Methodology for the differential diagnosis of complex data-set : A case study using data from routine CT-scan examinations Wijesinha, A.;Begg, C.B.;Funkenstein, H.H.;McNeil, B.J.