Dates, Times, and Locations:
There is no need for registration, simply drop by. We look forward to seeing you!
If you have any questions, or to borrow video tapes of past seminars, please contact Jae Lee:
Email: sjlee@mednet.ucla.edu
Telephone: 310-794-6270 Details:
HIV researchers regularly elicit detailed information on sexual behaviors from HIV+ and HIV-at-risk populations to assess HIV transmission behavior. We develop a multivariate hierarchical longitudinal model to analyze self-reported data on sexual behavior that can be customized to a range of data structures and substantive interests. We apply the model to evaluate behavior changes in partner count and HIV status disclosure over time for two distinct types of sexual partners. We use possibly zero-inflated Poisson random effects models for the longitudinal partner counts and a logistic random effects model for the disclosure data and fit the model in a Bayesian framework. The model can accommodate cross level predictors, within time and between time correlation and provides a template for more complex analyses that could include number of sexual acts and condom use. Our specific application is an analysis of the CLEAR (Choosing Life, Empowerment, Action, and Results) study which randomized HIV+ drug-abusing youths to a counseling intervention to redress transmission behavior.
*The research being presented in this seminar is co-authored by Lily Altstein.
Biography:
Robert Weiss, Ph.D. (University of Minnesota, Department of Statistics, 1989), has been a Professor of Biostatistics since 2002 at the UCLA School of Public Health, where he was previously Assistant Professor since 1990. He became a Fellow of the American Statistical Association in 2003. He is author of the book Modeling Longitudinal Data published by Springer in 2005 and is coauthor on 73 papers and counting. He has chaired or co-chaired 15 doctoral dissertations plus 3 ongoing dissertations. His statistical research interests are in hierarchical/multi-level models, Bayesian methods, longitudinal data, analysis of messy data, statistical graphics and statistical diagnostics. His subject matter interests run to applications in HIV research and analysis of human behavior.
***LIGHT REFRESHMENTS AT 3:00 PM *** |
