论文信息 - EM ALGORITHMS FOR GENERALIZED BRADLEY–TERRY MODELS

EM ALGORITHMS FOR GENERALIZED BRADLEY–TERRY MODELS

The Bradley–Terry model and its generalizations are popular models for paired (or multiple) comparisons of individuals or teams. The literature describes several approaches to maximum likelihood estimation of the parameters. Here we propose the use of the EM scheme, we study its convergence properties, and compare it with previous algorithms.

V. Csiszár | I. Kátai

[1] R. A. Bradley,et al. Rank Analysis of Incomplete Block Designs: I. The Method of Paired Comparisons , 1952 .

[2] P. V. Rao,et al. Ties in Paired-Comparison Experiments: A Generalization of the Bradley-Terry Model , 1967 .

[3] R. Plackett. The Analysis of Permutations , 1975 .

[4] Alan Agresti,et al. Categorical Data Analysis , 1991, International Encyclopedia of Statistical Science.

[5] G. McLachlan,et al. The EM algorithm and extensions , 1996 .

[6] D. Hunter. MM algorithms for generalized Bradley-Terry models , 2003 .

[7] Chih-Jen Lin,et al. Generalized Bradley-Terry Models and Multi-Class Probability Estimates , 2006, J. Mach. Learn. Res..

[8] John Guiver,et al. Bayesian inference for Plackett-Luce ranking models , 2009, ICML '09.