Comparing probabilistic predictive models applied to football
Applications
2020-06-16 v1
Abstract
We propose two Bayesian multinomial-Dirichlet models to predict the final outcome of football (soccer) matches and compare them to three well-known models regarding their predictive power. All the models predicted the full-time results of 1710 matches of the first division of the Brazilian football championship and the comparison used three proper scoring rules, the proportion of errors and a calibration assessment. We also provide a goodness of fit measure. Our results show that multinomial-Dirichlet models are not only competitive with standard approaches, but they are also well calibrated and present reasonable goodness of fit.
Keywords
Cite
@article{arxiv.1705.04356,
title = {Comparing probabilistic predictive models applied to football},
author = {Marcio A. Diniz and Rafael Izbicki and Danilo Lopes and Luis Ernesto Salasar},
journal= {arXiv preprint arXiv:1705.04356},
year = {2020}
}