|
Activity Number:
|
370
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 4, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Nonparametric Statistics
|
| Abstract - #305500 |
|
Title:
|
Multiple Imputation Based on Restricted Mean Models for Censored Survival Data
|
|
Author(s):
|
Lyrica Xiaohong Liu*+ and Susan Murray and Alexander Tsodikov
|
|
Companies:
|
University of Michigan and University of Michigan and University of Michigan
|
|
Address:
|
Department of Biostatistics, Ann Arbor, MI, 48109,
|
|
Keywords:
|
Multiple Imputation ; Pseudo Observation ; Restricted Mean Life
|
|
Abstract:
|
Most multiple imputation methods for censored survival data either ignore patient characteristics when imputing a likely event time, or place quite restrictive modeling assumptions on the survival distributions used for imputation. In this research, we propose a multiple imputation approach that directly imputes restricted lifetimes over the study period based on a model of the mean restricted life as a linear function of covariates. This method retains patient characteristics through the model on the mean structure when making imputation choices, but does not make assumptions on the shapes of hazards or survival functions. Simulation results show that the resulting model of mean restricted life gives more precise parameter estimates than a pseudo-value approach for fitting a similar model for the restricted mean, without making additional parametric assumptions.
|
