Modifying Bradley–Terry and other ranking models to allow ties

Author:

Baker Rose1,Scarf Philip2

Affiliation:

1. School of Business, University of Salford, The Crescent, Salford M5 4WT, UK

2. Cardiff Business School, Cardiff University, Aberconway Building, Colum Drive, Cardiff CF10 3EU, UK

Abstract

Abstract Models derived from distributions of order-statistics are useful for modelling ranked data. The well-known Bradley–Terry (BT) and Plackett–Luce (PL) models can be derived from the order statistics of the exponential distribution but cannot handle ties. However, ties often occur in sports, and the ability to accommodate them leads to more useful ranking models. In this paper, we use discrete distributions, principally the geometric distribution, to obtain modified BT and PL models and some others that allow tied ranks. Our methodology is introduced for some mathematically tractable and some less tractable distributions and is illustrated using test match cricket.

Publisher

Oxford University Press (OUP)

Subject

Applied Mathematics,Management Science and Operations Research,Strategy and Management,General Economics, Econometrics and Finance,Modeling and Simulation,Management Information Systems

Cited by 6 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Comparing machine learning algorithms by union-free generic depth;International Journal of Approximate Reasoning;2024-06

2. Modelling and Analysis of Rank Ordered Data with Ties via a Generalized Plackett-Luce Model;Bayesian Analysis;2024-01-01

3. Modelling player ratings in One Day International cricket using the Modified Duckworth-Lewis method;IMA Journal of Management Mathematics;2023-09-03

4. A unified theory for bivariate scores in possessive ball-sports: The case of handball;European Journal of Operational Research;2023-02

5. On skill and chance in sport;IMA Journal of Management Mathematics;2021-09-02

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3