|
Activity Number:
|
603
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 6, 2009 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Risk Analysis
|
| Abstract - #303790 |
|
Title:
|
Quantile Regression with Left-Truncated Semi-Competing Risks Data
|
|
Author(s):
|
Ruosha Li*+ and Limin Peng
|
|
Companies:
|
Emory University and Emory University
|
|
Address:
|
1518 Clifton Rd NE 3rd FL, Atlanta, GA, 30322,
|
|
Keywords:
|
Cumulative incidence function ; Estimating equation ; Left truncation ; Quantile regression ; Semi-competing Risks
|
|
Abstract:
|
We often encounter semi-competing risks data in biomedical studies where a terminating event censors a non-terminating event but not vice versa. In practice, left truncation on the terminating event may arise and can greatly complicate the assessment of covariate effects on the non-terminating event. We propose a quantile regression model based on cumulative incidence conditional quantiles. The proposed model provides meaningful interpretations of regression coefficients, and offers flexibility of accommodating varying covariate effects. We construct a set of unbiased monotone estimating equations that can be solved by existing statistical software. Inference procedures are developed accordingly. We establish asymptotic properties including uniform consistency and weak convergence. The performance of the proposed method would be illustrated by simulation studies and a real data analysis.
|
