|
Activity Number:
|
340
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 8, 2006 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #306915 |
|
Title:
|
Varying Coefficient Model with Measurement Error
|
|
Author(s):
|
Liang Li*+ and Tom Greene
|
|
Companies:
|
The Cleveland Clinic and The Cleveland Clinic
|
|
Address:
|
Department of Quantitative Health Sciences, Cleveland, OH, 44195,
|
|
Keywords:
|
measurement error ; varying coefficient model ; local polynomial ; chronic renal disease ; semi-parametric
|
|
Abstract:
|
We propose a semi-parametric partially varying coefficient model to study the creatinine clearance process in various populations. The coefficient of GFR is specified as a function of age to allow its effect to be age-dependent. The GFR measurements are assumed to be surrogates for the true GFR, their relationships modeled in an additive error model. We use local corrected score equations to estimate parameters and coefficient functions, and use an expected generalized cross-validation method to select the kernel bandwidth. The proposed method avoids making assumptions on the distributions of true GFR and residuals. An application of the proposed model reveals that there might be a consistent creatinine clearance process among healthy kidney donors and chronic renal disease patients.
|
