3/25/2016

Post date: Mar 28, 2016 4:55:44 PM

Title: Testing Independence of Bivariate Interval Censored Data By Modified Kendall’s Tau Statistic

Speaker: DoHwan Park, Department of Mathematics and Statistics, UMBC

[Abstract]

In this presentation, we discuss a nonparametric procedure to test independence of bivariate interval censored data; for both current status data and interval censored data. It is a score based modification of the Kendall’s tau statistic for bivariate interval censored data. The performance of the modified approach is illustrated by simulation studies and application to the AIDS study. We compare our method to alternative approaches such as two-stage estimation method Sun et al. (2006) and multiple imputation method by Betensky and Finkelstein (1999b). It is joint work with Y. Kim and J. Lim.