English

Sparse Poisson Regression with Penalized Weighted Score Function

Statistics Theory 2017-03-14 v1 Statistics Theory

Abstract

We proposed a new penalized method in this paper to solve sparse Poisson Regression problems. Being different from 1\ell_1 penalized log-likelihood estimation, our new method can be viewed as penalized weighted score function method. We show that under mild conditions, our estimator is 1\ell_1 consistent and the tuning parameter can be pre-specified, which shares the same good property of the square-root Lasso.

Keywords

Cite

@article{arxiv.1703.03965,
  title  = {Sparse Poisson Regression with Penalized Weighted Score Function},
  author = {Jinzhu Jia and Fang Xie and Lihu Xu},
  journal= {arXiv preprint arXiv:1703.03965},
  year   = {2017}
}
R2 v1 2026-06-22T18:43:02.008Z