BTdecayLasso

CRANBTdecayLassoMaRDI QIDQ149775FDOQ149775

Bradley-Terry Model with Exponential Time Decayed Log-Likelihood and Adaptive Lasso

Jinfeng Xu, Yunpeng Zhou

Last update: 7 December 2023

Copyright license: GNU General Public License, version 3.0, GNU General Public License, version 2.0

Software version identifier: 0.1.0, 0.1.1

Description

We utilize the Bradley-Terry Model to estimate the abilities of teams using paired comparison data. For dynamic approximation of current rankings, we employ the Exponential Decayed Log-likelihood function, and we also apply the Lasso penalty for variance reduction and grouping. The main algorithm applies the Augmented Lagrangian Method described by Masarotto and Varin (2012) <doi:10.1214/12-AOAS581>.

Cites work

• The ranking lasso and its application to sport tournaments

This page was built for software: BTdecayLasso