科学研究
报告题目:

Sparse Poisson Regression with Penalized Weighted Score Function

报告人:

贾金柱 教授(北京大学)

报告时间:

报告地点:

数学院二楼报告厅

报告摘要:

We proposed a new penalized method in this paper to solve sparse Poisson Regression problems. Being different from l1 penalized log- likelihood estimation, our new method can be viewed as penalized weighted score function method, which produces a tuning-free feature. We show that under mild conditions, our estimator is l1 consistent and the tuning parameter can be pre-specified, which shares the same good property of the square-root Lasso.