"Annie" Peiyong Qu- Colloquium Speaker

Professor, Department of Statistics, University of Illinois at Urbana-Champaign
Date: 
Thursday, April 2, 2015 - 3:30pm
Colloquium Title: 
Weak Signal Identification and Inference in Penalized Model Selection
Location: 
Reception at 3:00 p.m. in 241 SH / Talk at 3:30 in 61 SH

Abstract:

Weak signal identification and inference are very important in the area of penalized model selection, yet they are under-developed and not well-studied.  Existing inference procedures for  penalized estimators are mainly focused on strong signals. In this paper, we propose an identification procedure for weak signals in finite samples,  and provide  a transition phase in-between noise and strong signal strengths. We also introduce a  new two-step inferential method  to construct  better confidence intervals  for  the identified weak signals. Both theory and numerical studies indicate that the proposed method  leads to better confidence coverage  for weak signals, compared with those using asymptotic inference. In addition, the proposed  method  outperforms the  perturbation  and bootstrap resampling approaches. We  illustrate our method for HIV antiretroviral drug susceptibility data to identify genetic mutations associated with HIV drug resistance. This is joint work with Peibei Shi.

poster