Mallows Ranking Models: Maximum Likelihood Estimate and Regeneration
Statistics Theory
2019-06-18 v2 Probability
Methodology
Statistics Theory
Abstract
This paper is concerned with various Mallows ranking models. We study the statistical properties of the MLE of Mallows' model. We also make connections of various Mallows ranking models, encompassing recent progress in mathematics. Motivated by the infinite top- ranking model, we propose an algorithm to select the model size automatically. The key idea relies on the renewal property of such an infinite random permutation. Our algorithm shows good performance on several data sets.
Keywords
Cite
@article{arxiv.1808.08507,
title = {Mallows Ranking Models: Maximum Likelihood Estimate and Regeneration},
author = {Wenpin Tang},
journal= {arXiv preprint arXiv:1808.08507},
year = {2019}
}
Comments
10 pages, 2 figures, 5 tables. This paper is published by http://proceedings.mlr.press/v97/tang19a.html