VSE-FACULTY-L Archives

March 2014

VSE-FACULTY-L@LISTSERV.GMU.EDU

Options: Use Monospaced Font
Show HTML Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Lisa Nolder <[log in to unmask]>
Reply To:
Lisa Nolder <[log in to unmask]>
Date:
Mon, 31 Mar 2014 08:55:34 -0400
Content-Type:
multipart/alternative
Parts/Attachments:
text/plain (2005 bytes) , text/html (25 kB)
*_Notice and Invitation_*
Oral Defense of Doctoral Dissertation

Department of Statistics
Volgenau School of Engineering, George Mason University

*Mengdie Yuan*

Bachelor of Science, Statistics, University of Science and Technology of 
China, 2009*
*Master of Science, Statistical Science, George Mason University, 2011*

*

*Semiparametric Regression Analysis of Survival and Longitudinal Data*

Monday, April 28^th , 2014

10:30 AM to 12:30 PM
The Nguyen Engineering Building, Room 3507


All are invited to attend.

*_Committee_*
Guoqing Diao, Chair
Jacqueline Hughes-Oliver

Anand Vidyashankar
Ali Weinstein


*_Abstract_*

The proportional odds model is a popular model in survival analysis. The 
assumption of constant odds ratios over time in the proportional odds 
model, however, is often violated in many applications. We propose a 
novel semiparametric general odds ratio model for the analysis of 
right-censored survival data. The proposed model incorporates the 
short-term and long-term covariate effects on the failure time data, and 
includes the proportional odds model as a special case. We derive 
efficient likelihood-based inference procedures and establish the large 
sample properties of the proposed nonparametric maximum likelihood 
estimators. Extensive simulation studies demonstrate the proposed 
methods perform well in practical settings. An application to a breast 
cancer study is provided.

Secondly, we study the generalized linear models with an unknown link 
function for both data with independent observations and longitudinal 
data. We propose sieve maximum likelihood estimation procedures for both 
fixed effects models and random effects models by using B-splines. We 
establish the consistency and asymptotic normality of the proposed sieve 
maximum likelihood estimators.Extensive simulation studies along with an 
application to anepileptic study are provided to evaluate the 
finite-sample performance of the proposed methods.

A copy of this doctoral dissertation will be on reserve at the Johnson 
Center Library.



ATOM RSS1 RSS2