Activity Number:
|
192
- Contributed Poster Presentations: Section on Statistics and the Environment
|
Type:
|
Contributed
|
Date/Time:
|
Monday, July 30, 2018 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Section on Statistics and the Environment
|
Abstract #329799
|
|
Title:
|
Different Methods and Comparisons Dealing with Censored Count Data
|
Author(s):
|
Xiao Yu* and Lung-Chang Chien and Kai Zhang
|
Companies:
|
University of Texas Health Science Center at Houson and University of Nevada, Las Vegas and University of Texas Health Science Center at Houson
|
Keywords:
|
censored count data;
Mulitple Imputation;
Small Area Estimation;
Censored Poisson regression model
|
Abstract:
|
Count data are commonly used to report counting statistics of diverse health outcomes. However, some data are marked on purpose to avoid leaking information to identify individuals when population sizes are small. The situation hinders the further use from those data in public health research. We therefore considered three methods to deal with censored count data: 1)multiple imputation(MI) method; 2)small area estimation method; 3)censored Poisson regression method. A series of simulations results in that the censored Poisson regression method conducted the closest estimates to the true values (with the relative error = 0.5%), and MI had the worst results (with relative error=27.5%) under the censored proportion 7.9%. After comparing the results under the censored proportion by 10.6% and 33.6%, the censored Poisson regression method still shows a smaller relative error than the other two methods. We also applied this these three methods to evaluate the health effects of air pollution and weather extremes on hospital admissions in Greater Houston Area. To sum up, applying the censored Poisson regression method can conduct the best performance when dealing with censored count data.
|
Authors who are presenting talks have a * after their name.