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 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 consistent and the tuning parameter can be pre-specified, which shares the same good property of the square-root Lasso.
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}
}