• Journal of the Korean Data and Information Science Society
• /
• v.24 no.5
• /
• pp.1089-1099
• /
• 2013

The Bradley-Terry model is widely used for analysis of pairwise preference data. We explain that the popularity of Bradley-Terry model is gained due to not only easy computation but also some nice asymptotic properties when the model is misspecified. For information retrieval required to analyze big ranking data, we propose to use a pseudo likelihood based on the Bradley-Terry model even when the true model is different from the Bradley-Terry model. We justify using the Bradley-Terry model by proving that the estimated ranking based on the proposed pseudo likelihood is consistent when the true model belongs to the class of Thurstone models, which is much bigger than the Bradley-Terry model.

• The Korean Journal of Applied Statistics
• /
• v.24 no.5
• /
• pp.915-930
• /
• 2011

There are a variety of methods to model game results and many methods exist for the case of paired comparison data. Among them, the Bradley-Terry model is the most widely used to derive a latent preference scale from paired comparison data. It has been applied in a variety of fields in psychology and related disciplines. We applied this model to the data of Korean Baseball League. It shows that the loglinear Bradley-Terry model of defensive rate and save is optimal in terms of AIC. Also some categorical characteristics, such as east team and west team, existence of golden glove winning players, team(s) with seasonal pitching leader, and team(s) with home advantage, influenced the game result significantly. As a result, the suggested models can be further utilized to predict future game results.

• The Korean Journal of Applied Statistics
• /
• v.20 no.2
• /
• pp.205-218
• /
• 2007

It is our greatest concern of Korean team to enter round of 16. The past football results are the most important data for making a prediction. And we know that the home advantage is also considerable factor and there are many unobservable factors. However, there are few matches between the participants and even not the results for some nations. To overcome this difficulty, we model the network of results and consider other factors. We predict 2006 Germany World Cup results using modified the Bradley-Terry model.

• The Korean Journal of Applied Statistics
• /
• v.32 no.6
• /
• pp.783-794
• /
• 2019

The current ranking system of the Korean Baduk Association is based on the Elo rating system, which is widely used in the field of chess. Despite the 6.5 point dum (penalty) as compensation for playing as White, many Baduk players still prefer to playing as Black due to Black's higher winning percentage. In this paper, we present the ranking of Baduk players based on the Bradley-Terry model and address the advantage of playing as Black. We compare the ranking from our model with rankings from the Korean Baduk Association.

• Journal of the Korean Statistical Society
• /
• v.33 no.1
• /
• pp.129-147
• /
• 2004

For square contingency tables with nominal categories, this paper proposes a measure to represent the degree of departure from the quasi-symmetry (QS) model and the Bradley-Terry (BT) model. The measure proposed is expressed by using the Cressie and Read (1984)'s power-divergence or Patil and Taillie (1982)'s diversity index. The measure lies between 0 and 1, and it is useful for comparing the degree of departure from QS or BT in several tables.

• Journal of the Korean Data and Information Science Society
• /
• v.21 no.6
• /
• pp.1137-1146
• /
• 2010

There are a lot of methods to predict the result of a game and many forecasting researches have been studied. Among many methods, if a statistical model including some realistic random variables is used to forecast, more accurate prediction could be expected than any others. In this work, Bradley-Terry model is considered to predict results of 2010 South Africa World Cup games via paired comparison method. This prediction model includes some random variables which affect the results of games. The worth parameters for each country in this model are convergence values obtained by using Newton-Raphson algorithm. With this model, we can forecast top 16 among 32 countries and up to who will win the victory. Final results of 2010 South Africa World Cup games are compared with this prediction and discuss further works.