|
Activity Number:
|
116
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
| Abstract - #305765 |
|
Title:
|
Using Empirical Likelihood Confidence Regions to Establish Spirometry Limits
|
|
Author(s):
|
Nancy L. Glenn *+
|
|
Companies:
|
Texas Southern University
|
|
Address:
|
874 Yorkchester #121, Houston, TX, 77079,
|
|
Keywords:
|
Empirical Likelihood ; Spirometry
|
|
Abstract:
|
Spirometry reference values represent the volume of air entering and leaving the lungs, which aids in pulmonary diagnoses. In typical spirometry reference data sets, limits established to determine healthy pulmonary function are given as confidence intervals that use the lower bound or lower limit of normality method. However, these methods are based on the assumption that the scatter around the predicted value is normally distributed. In practice, this assumption is often violated. We construct empirical likelihood confidence bands to establish normal limits for spirometry data. The empirical likelihood is a nonparametric likelihood that does not rely on the assumption of normality or other distributional assumptions. This approach uses the maximum likelihood which is one the most popular methods for deriving statistical estimators.
|
