Penalized empirical likelihood over decentralized networks
Authors
Research Topics
Paper Information
-
Journal:
Journal of the Royal Statistical Society Series B -
DOI:
10.1093/jrsssb/qkag042 -
Published:
February 10, 2026 -
Added to Tracker:
Feb 11, 2026
Abstract
Abstract Empirical likelihood encounters serious computational challenges when applied to massive datasets or multiple data sources distributed across decentralized networks. This paper proposes a constrained empirical likelihood framework for decentralized networks, utilizing a novel penalization technique to obtain a penalized empirical log-likelihood. The resulting empirical log-likelihood ratio statistic is proved to be asymptotically standard chi-squared even for a divergent machine number. However, the optimization problem with the fused penalty is still hard to solve in the decentralized distributed network due to the coupling structure. To address the problem, two novel algorithms are developed to solve the optimization problem in a decentralized manner, with established convergence properties and linear convergence for the second algorithm in specific network structures. The methods are validated through simulations and real data analyses of census income and Ford gobike datasets.
Author Details
Jinye Du
AuthorQihua Wang
AuthorResearch Topics & Keywords
High-Dimensional Statistics
Research AreaCitation Information
APA Format
Jinye Du
&
Qihua Wang
(2026)
.
Penalized empirical likelihood over decentralized networks.
Journal of the Royal Statistical Society Series B
, 10.1093/jrsssb/qkag042.
BibTeX Format
@article{paper897,
title = { Penalized empirical likelihood over decentralized networks },
author = {
Jinye Du
and Qihua Wang
},
journal = { Journal of the Royal Statistical Society Series B },
year = { 2026 },
doi = { 10.1093/jrsssb/qkag042 },
url = { https://doi.org/10.1093/jrsssb/qkag042 }
}