A common problem faced in statistical inference is drawing conclusions from
paired comparisons, in which two objects compete and one is declared the
victor. A probabilistic approach to such a problem is the Bradley-Terry model,
first studied by Zermelo in 1929 and rediscovered by Bradley and Terry in 1952.
One obvious area of application for such a model is sporting events, and in
particular Major League Baseball. With this in mind, we describe a hierarchical
Bayesian version of Bradley-Terry suitable for use in ranking and prediction
problems, and compare results from these application domains to standard
maximum likelihood approaches. Our Bayesian methods outperform the MLE-based
analogues, while being simple to construct, implement, and interpret.
Description
Hierarchical Bayesian Bradley-Terry for Applications in Major League
Baseball
%0 Generic
%1 phelan2017hierarchical
%A Phelan, Gabriel C.
%A Whelan, John T.
%D 2017
%K bradley-terry sport
%T Hierarchical Bayesian Bradley-Terry for Applications in Major League
Baseball
%U http://arxiv.org/abs/1712.05879
%X A common problem faced in statistical inference is drawing conclusions from
paired comparisons, in which two objects compete and one is declared the
victor. A probabilistic approach to such a problem is the Bradley-Terry model,
first studied by Zermelo in 1929 and rediscovered by Bradley and Terry in 1952.
One obvious area of application for such a model is sporting events, and in
particular Major League Baseball. With this in mind, we describe a hierarchical
Bayesian version of Bradley-Terry suitable for use in ranking and prediction
problems, and compare results from these application domains to standard
maximum likelihood approaches. Our Bayesian methods outperform the MLE-based
analogues, while being simple to construct, implement, and interpret.
@misc{phelan2017hierarchical,
abstract = {A common problem faced in statistical inference is drawing conclusions from
paired comparisons, in which two objects compete and one is declared the
victor. A probabilistic approach to such a problem is the Bradley-Terry model,
first studied by Zermelo in 1929 and rediscovered by Bradley and Terry in 1952.
One obvious area of application for such a model is sporting events, and in
particular Major League Baseball. With this in mind, we describe a hierarchical
Bayesian version of Bradley-Terry suitable for use in ranking and prediction
problems, and compare results from these application domains to standard
maximum likelihood approaches. Our Bayesian methods outperform the MLE-based
analogues, while being simple to construct, implement, and interpret.},
added-at = {2018-01-08T13:01:45.000+0100},
author = {Phelan, Gabriel C. and Whelan, John T.},
biburl = {https://www.bibsonomy.org/bibtex/20de374638407dfcdc45057b3668113cd/dfirth},
description = {Hierarchical Bayesian Bradley-Terry for Applications in Major League
Baseball},
interhash = {9ae4aa65125833ca43274fe07a5cf898},
intrahash = {0de374638407dfcdc45057b3668113cd},
keywords = {bradley-terry sport},
note = {cite arxiv:1712.05879},
timestamp = {2018-01-08T13:01:45.000+0100},
title = {Hierarchical Bayesian Bradley-Terry for Applications in Major League
Baseball},
url = {http://arxiv.org/abs/1712.05879},
year = 2017
}