Colloquium Talk

GEORGETOWN UNIVERSITY

Department of Mathematics & Statistics

High-Dimensional Inference of Ordinal Data with Medical Applications

Feiran Jiao

Office of Biostatistics, Center of Drug Evaluation and Research, U.S. Food and Drug Administration

Friday, March 24th, 2017 @ 3:15pm

St. Mary’s Hall 326

 

Abstract:  

In this work, we develop new statistical methods for variable selection in a high-dimensional cumulative link regression model with an ordinal response. Our study is partly motivated by the needs for exploring the association structure between disease phenotype and high-dimensional medical covariates. We proposed a penalized maximum likelihood approach with a composite bridge penalty to solve the bi-level selection problem in a cumulative link model. A minorization-maximization (MM) algorithm was developed for implementing the proposed method, which is specific to each of the 4 link functions, namely, probit, logistic, Cauchy and complementary loglog links. The proposed approach is shown to enjoy a number of desirable theoretical properties including bi-level selection consistency and oracle properties, under suitable regularity conditions. Simulations demonstrate that the proposed method enjoys good empirical performance. We illustrated the proposed methods with a real medical application.

 

Friday, March 24 at 3:15pm

St. Mary's Hall, 326
3700 Reservoir Road, N.W., Washington

Event Type

Academic Events

Departments

Georgetown College, Mathematics and Statistics

Event Contact Name

Nate Strawn

Subscribe
Google Calendar iCal Outlook

Recent Activity