|
Activity Number:
|
337
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #309638 |
|
Title:
|
Bayesian Sample Size Calculation for Counted Data with Excess Zeroes
|
|
Author(s):
|
Chunyao Feng*+ and John W. Seaman, Jr. and James Stamey
|
|
Companies:
|
Amgen Inc. and Baylor University and Baylor University
|
|
Address:
|
One Amgen Center Drive, Thousand Oaks, CA, 91320,
|
|
Keywords:
|
Bayesian ; Sample Size ; Negative Binomial ; Zero-inflated Poisson ; Generalized Poisson ; Model
|
|
Abstract:
|
Hospitalization is the largest portion of the total cost of patients to the health services. Interest lies in comparing the rate of hospitalization for two groups of patients. We consider a Bayesian method to determine the required sample size to test equality of these rates with a pre-specified power. Generalized Poisson (GP), Negative Binomial (NB) models, Zero-inflated Poisson (ZIP) and Zero-inflated Negative Binomial (ZINB) models are considered as the candidates to model the rate of hospitalization. In addition, we compare the performance of these models via Bayesian model comparison techniques. Finally, a sensitivity analysis is performed to test the robustness of this methodology.
|
