English

Communication-Efficient Distributed Estimator for Generalized Linear Models with a Diverging Number of Covariates

Methodology 2020-08-14 v2 Distributed, Parallel, and Cluster Computing Machine Learning Machine Learning

Abstract

Distributed statistical inference has recently attracted immense attention. The asymptotic efficiency of the maximum likelihood estimator (MLE), the one-step MLE, and the aggregated estimating equation estimator are established for generalized linear models under the "large nn, diverging pnp_n" framework, where the dimension of the covariates pnp_n grows to infinity at a polynomial rate o(nα)o(n^\alpha) for some 0<α<10<\alpha<1. Then a novel method is proposed to obtain an asymptotically efficient estimator for large-scale distributed data by two rounds of communication. In this novel method, the assumption on the number of servers is more relaxed and thus practical for real-world applications. Simulations and a case study demonstrate the satisfactory finite-sample performance of the proposed estimators.

Keywords

Cite

@article{arxiv.2001.06194,
  title  = {Communication-Efficient Distributed Estimator for Generalized Linear Models with a Diverging Number of Covariates},
  author = {Ping Zhou and Zhen Yu and Jingyi Ma and Maozai Tian and Ye Fan},
  journal= {arXiv preprint arXiv:2001.06194},
  year   = {2020}
}
R2 v1 2026-06-23T13:13:44.933Z