Academy of Mathematics and Systems Science, CAS Colloquia & Seminars
Speaker:
Prof. Runze Li, Penn State University
Inviter:
陈敏
Title:
Test of high dimensional mean vector
Time & Venue:
2017.12.8 10:00-11:00 N613
Abstract:
This talk will introduce a new test for high dimensional mean vector based on empirical likelihood method. When the dimension of the mean vector is greater than the sample size, the empirical likelihood test is not well defined. We will propose a modification of the empirical likelihood test for high dimensional mean vector. The modified empirical likelihood test has a closed form, and therefore it is easy to implement. We derive the asymptotic distribution of the test statistic under both the null hypothesis and alternative hypothesis. Numerical results demonstrate the proposed test may outperform the existing ones under certain scenarios.