|
Activity Number:
|
59
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 3, 2003 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #300643 |
|
Title:
|
A Smooth Estimate of the Bivariate Survival Density in the Presence of Left-, Right-, and Interval-Censored Data
|
|
Author(s):
|
Kris Bogaerts*+ and Emmanuel M. E. H. Lesaffre
|
|
Companies:
|
Katholieke Universiteit Leuven and Biostatistical Centre
|
|
Address:
|
Biostatistical Centre, Leuven, , 3000, Belgium
|
|
Keywords:
|
interval-censored data ; smoothing ; survival functions
|
|
Abstract:
|
In a recent longitudinal study (Signal-Tandmobiel) emergence times of 28 permanent teeth were recorded. Due to the study design, the emergence times were left-, right-, or interval-censored. The dentists are interested in evaluating the association pattern of the emergence times among the teeth. This involves fitting the bivariate distributions of the emergence times. Parametric models are relatively easy to fit. But to asses the goodness of fit of these models, bivariate exploration is required. This can be achieved by comparing the fit to the nonparametric maximum likelihood estimate. However, the nonparametric maximum likelihood estimate is not always unique. Therefore we propose a comparison to a bivariate smooth estimate of the density of emergence times. The approach is based on using B-splines and a penalized likelihood approach. A simulation study and an application to the Signal-Tandmobiel study show the performance of the method. In addition, the mathematical properties of the method will be given.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2003 program |
