Abstract #301496

The views expressed here are those of the individual authors
and not necessarily those of the ASA or its board, officers, or staff.

Back to main JSM 2002 Program page
JSM 2002 Abstract #301496
Activity Number: 288
Type: Contributed
Date/Time: Wednesday, August 14, 2002 : 8:30 AM to 10:20 AM
Sponsor: Biometrics Section*
Abstract - #301496
Title: Nonparametric Tree-tructured Modeling for Interval-censored Survival Data
Author(s): Yanming Yin*+ and Stewart Anderson
Affiliation(s): Novartis, Inc. and University of Pittsburgh
Address: 22C Troy Drive, Springfield, New Jersey, 07081, USA
Keywords: Interval-censored ; survival ; EM algorithm ; Nonparametric ; tree-structured modeling
Abstract:

A recursive partitioning algorithm is proposed for interval-censored time to event data. In this method, we make no assumption about the underlying failure distribution. To obtain a splitting criterion, we assume that each individual's event time falls into one of a discrete number of intervals. Consequently, by conditioning on the number of intervals, one can assume that the probability of being in a particular interval follows a multinomial distribution and thus obtains a log-likelihood function across individuals. Turnbull's EM algorithm (1976, Journal of the Royal Statistical Society, Series B 38, 290-295) is then used to estimate the associated probabilities.

In this method, pruning and cross-validation algorithms are similar to those of the Classification and Regression Tree (CART). The performance of our method is evaluated through simulation. In addition, an example is given to illustrate our method.

  • 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 2002 program

JSM 2002

For information, contact meetings@amstat.org or phone (703) 684-1221.

If you have questions about the Continuing Education program, please contact the Education Department.

Revised March 2002