A Penalized Maximum Likelihood Approach for the Ranking of College Football Teams Independent of Victory Margins

Mease, David

doi:10.1198/0003130032396

Cited by 55 publications

(51 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A problem that arises from this assumption is that if a given item i has won all pairwise contests, the likelihood becomes larger as sni becomes larger. It follows that a maximum likelihood estimate for sni is ∞ [24]. As a consequence the model will always produce a tie amongst all undefeated items.…”

Section: Score-based Methodsmentioning

confidence: 99%

CRF framework for supervised preference aggregation

Volkovs

Zemel

2013

Proceedings of the 22nd ACM International Conference on Information &Amp; Knowledge Management

View full text Add to dashboard Cite

We develop a flexible Conditional Random Field framework for supervised preference aggregation, which combines preferences from multiple experts over items to form a distribution over rankings. The distribution is based on an energy comprised of unary and pairwise potentials allowing us to effectively capture correlations between both items and experts. We describe procedures for learning in this model and demonstrate that inference can be done much more efficiently than in analogous models. Experiments on benchmark tasks demonstrate significant performance gains over existing rank aggregation methods.

show abstract

Section: Score-based Methodsmentioning

confidence: 99%

CRF framework for supervised preference aggregation

Volkovs

Zemel

2013

Proceedings of the 22nd ACM International Conference on Information &Amp; Knowledge Management

View full text Add to dashboard Cite

show abstract

“…Such techniques were developed in competitive sport judging/scoring systems [18,28,34], but to the best of our knowledge have never been attempted in Thurstone scaling. We show that the results of maximum likelihood estimation can be presented as a probit or logit regression model, and a software designed for binary regression can be used for Thurstone scaling.…”

Section: Introductionmentioning

confidence: 99%

Thurstone scaling via binary response regression

Lipovetsky

Conklin

2004

Statistical Methodology

View full text Add to dashboard Cite

“…Further, ad hoc parameters are not justifiable in a statistical sense -if they were, they wouldn't be ad hoc. It should be noted that since this trouble arises solely because some teams are unbeaten (or winless), there are approaches, such as penalized maximum likelihood (see Mease [18]), which overcome this difficulty but they are often subjective and thus decisions made by the analyst can affect the rankings.…”

Section: Win/loss Modelsmentioning

confidence: 99%

Hybrid Paired Comparison Analysis, with Applications to the Ranking of College Football Teams

Annis¹,

Craig²

2005

Journal of Quantitative Analysis in Sports

View full text Add to dashboard Cite

Existing paired comparison models used for ranking football teams primarily focus on either wins and losses or points scored (either via each team's total or a margin of victory). While reasonable, each approach fails to produce satisfactory rankings in frequently arising situations due to its ignorance of additional data. We propose a new, hybrid model incorporating both wins and constituent scores and show that it outperforms its competitors and is robust against model mis-specification based on a series of simulation studies. We conclude by illustrating the method using the 2003-04 and 2004-05 college football seasons.

show abstract

A Penalized Maximum Likelihood Approach for the Ranking of College Football Teams Independent of Victory Margins

Cited by 55 publications

References 8 publications

CRF framework for supervised preference aggregation

CRF framework for supervised preference aggregation

Thurstone scaling via binary response regression

Hybrid Paired Comparison Analysis, with Applications to the Ranking of College Football Teams

Contact Info

Product

Resources

About