Model fitting using partially ranked data

The importance of models for complete ranking data is well-established in the literature. Partial rankings, on the other hand, arise naturally when the set of objects to be ranked is relatively large. Partial rankings give rise to classes of compatible order preserving complete rankings. In this art...

Full description

Saved in:
Bibliographic Details
Published in:Journal of nonparametric statistics 2023-07, Vol.35 (3), p.587-600
Main Authors: Alvo, Mayer, Duan, Xiuwen
Format: Article
Language:English
Subjects:
Algorithms
compatibility
Data analysis
EM algorithm
incomplete rankings
Mallows exponential model
Ranking systems
Statistical analysis
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The importance of models for complete ranking data is well-established in the literature. Partial rankings, on the other hand, arise naturally when the set of objects to be ranked is relatively large. Partial rankings give rise to classes of compatible order preserving complete rankings. In this article, we define an exponential model for complete rankings and calibrate it on the basis of a random sample of partial rankings data. We appeal to the EM algorithm. The approach is illustrated in some simulations and in real data.
ISSN:1048-5252
1029-0311
DOI:10.1080/10485252.2023.2176180