Presented by: 
Liqun Diao, University of Rochester
Date: 
Fri 23 Aug, 11:00 am - 12:00 pm
Venue: 
67-142

In this talk, I will introduce my research on statistical modeling of data in the
studies of chronic diseases which involve modeling dependencies between life history
processes using copula functions. Specifically, I will cover the following three topics:
(1) a copula-based framework is considered for jointly analyzing damage progression
of several organ systems by assuming that the joint survival function of absorption
transition times is governed by a multivariate copula function; (2) a copula model
is adopted for marked point process using the idea of governing associated marks
and event times with copula functions; (3) a construction based on vine copula
model is considered for the recurrent events with non-ignorable durations to reflect
multiple dependencies between the episode onset times and the duration of the episodes.