|
Activity Number:
|
369
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 6, 2008 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #302367 |
|
Title:
|
Semiparametric Inference of Linear Transformation Models with Length-Biased Censored Data
|
|
Author(s):
|
Jane Paik*+ and Zhiliang Ying
|
|
Companies:
|
Columbia University and Columbia University
|
|
Address:
|
1255 Amsterdam Avenue, New York, NY, 10027,
|
|
Keywords:
|
Semiparametric linear transformation model ; biased sampling ; censoring
|
|
Abstract:
|
We propose estimating equations for the simultaneous estimation of regression parameters and the transformation function in semiparametric linear transformation models when dealing with biased sampling in censored data. Existing estimation procedures for censored data using linear transformation models yield biased estimators for regression parameters of interest. Our approach is motivated by the unified estimation procedure proposed by Chen et al. (2002). The proposed estimators for the regression parameters are proven to be consistent and asymptotically normal. The variance-covariance matrix has a closed form which can be consistently estimated. The finite sample performances under various scenarios are assessed through simulation studies, which indicate that the proposed estimators give negligible bias and correct coverage probabilities. The method is also applied to a real data set.
|
