|
Activity Number:
|
511
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #305077 |
|
Title:
|
A Hidden Markov Model for Zero-Inflated Poisson Counts with an Application to Cocaine Use Data
|
|
Author(s):
|
Stacia M. DeSantis*+ and Dipankar Bandyopadhyay
|
|
Companies:
|
Medical University of South Carolina and Medical University of South Carolina
|
|
Address:
|
150 Cannon St , Charleston, SC, 29425,
|
|
Keywords:
|
Bayesian ; hidden Markov model ; Poisson ; drug abuse
|
|
Abstract:
|
Paradigms for substance abuse cue-reactivity research involve short term pharmacological or stressful stimulation designed to elicit stress and craving responses in cocaine-dependent subjects. It is unclear as to whether psychosocial stress induced from participation in such studies increases drug-seeking behavior. We propose a 2-state Hidden Markov model to model the number of cocaine abuses per week before and after participation in the study. The latent states correspond to "abuse"' or "no abuse." To account for a preponderance of zeros, we assume a zero-inflated Poisson process for the count data. Transition probabilities depend on the prior week's state, fixed demographic variables, and time-varying covariates. We adopt a Bayesian approach to model fitting, and use the conditional predictive ordinate statistic to compare our model to other models for longitudinal count data.
|
