American Statistical Association
Hidden Markov models (HMM) are useful for modeling longitudinal self-reports of drug or alcohol use as a function of treatments and time dependent variables. We develop a HMM for zero inflated Poisson counts in order to assess the impact of time dependent interventions on transitioning among substance use and drinking states derived from a zero-inflated measure. The model is fit under a Bayesian paradigm and the conditional predictive ordinate is used for model selection. We reanalyze a small study of cocaine dependent participants and The Combined Pharmacotherapies and Behavioral Interventions for Alcohol Dependence (COMBINE) trial.
Stacia DeSantis is an Assistant Professor in the Division of Biostatistics and Epidemiology in the Department of Medicine at the Medical University of South Carolina. Her research interests include statistical methods in Psychiatry and Neuroscience and Bayesian biostatistics.
|Date:||Tuesday, April 10, 2012|
|Time:||3:30 - 4:30 P.M.|
New York State Psychiatric Institute
1051 Riverside Drive
6th Floor Multipurpose Room (6602)
New York, New York