|
Activity Number:
|
287
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 4, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Teaching of Statistics in the Health Sciences
|
| Abstract - #304076 |
|
Title:
|
Bias-Corrected Estimation for Intraclass Correlation Coefficient
|
|
Author(s):
|
Eshetu G. Atenafu*+ and Joseph Beyene and Jemila S. Hamid
|
|
Companies:
|
Hospital for Sick Children and University of Toronto and Hospital for Sick Children
|
|
Address:
|
555 University Ave., Toronto, ON, M5G 1X8, Canada
|
|
Keywords:
|
Bias ; Intra-Class Correlation ; ANOVA ; Taylor Series ; Simulation
|
|
Abstract:
|
A bias-corrected estimator is proposed for the intra-class correlation coefficient (ICC) using second order Taylor series expansion. In this presentation we consider a particular type of ICC from one-way random effects analysis of variance model, although our approach can be adapted to construct bias-corrected estimators for other types of ICCs in general. A simulation study is performed to compare the proposed estimator with the conventional estimator, both on normal and non-normal data settings. A marked bias reduction is obtained using the newly proposed estimator in both scenarios. In particular, the estimator that is bias-corrected outperforms the analytic estimator in a non-normal setting producing estimates that are very close to the true ICC values.
|
