Models for Paired Comparison Data: A Review with Emphasis on Dependent Data

Abstract

Thurstonian and Bradley-Terry models are the most commonly applied models in the analysis of paired comparison data. Since their introduction, numerous developments have been proposed in different areas. This paper provides an updated overview of these extensions, including how to account for object-and subject-specific covariates and how to deal with ordinal paired comparison data. Special emphasis is given to models for dependent comparisons. Although these models are more realistic, their use is complicated by numerical difficulties. We therefore concentrate on implementation issues. In particular, a pairwise likelihood approach is explored for models for dependent paired comparison data, and a simulation study is carried out to compare the performance of maximum pairwise likelihood with other limited information estimation methods. The methodology is illustrated throughout using a real data set about university paired comparisons performed by students.

Characterizing and Predicting Early Reviewers for Effective Product Marketing on E-Commerce Websites

“…Comparison-based preference has been studied for several decades [31], [35], [36], and a survey of the classic approaches and methods was given in [37]. By modeling comparisonbased preference, we can essentially perform any ranking task.…”

Elo Ratings and the Sports Model: A Neglected Topic in Applied Probability?

“…The basic statistics theory (MLEs, confidence intervals, hypothesis tests, goodness-of-fit tests) of that model is treated in Chapter 4 of David (1988). See Cattelan (2012) for a recent survey. To use the model, we need to specify how matches are scheduled.…”

Bayesian Analysis of the Rayleigh Paired Comparison Model under Loss Functions using Informative Prior

“…= E( ), is the Bayes estimator, and 2 ( ) = E( 2 ) E( ) 2 , is the Bayes posterior risk. Table 5 The Bayes estimators under loss function, L 3 , have overall minimum risk, more than those under L 1 and L 2 using both the Conjugate and Dirichlet priors.…”

“…Three methods for elicitation are recommended for the case of two treatments and one method for the general case. Cattelan [2] presented the extensions in the Thurstonian and Bradley-Terry models on how to account for object-and subject-speci c covariates. Models of dependent comparisons were also considered.…”