Speaker: Emily Berg, Associate Professor, Department of Statistics, Iowa State University

Title: A Multivariate Bayesian Hierarchical Model for Small Area Estimation of Criminal Victimization Rates in Domains Defined by Age and Gender

Abstract:  The National Crime and Victimization Survey (NCVS) gathers information on criminal victimizations for individuals in a representative sample of United States households.  The NCVS provides authoritative data on the rates of many types of violent crimes, including simple assault, robbery, and aggravated assault.   The NCVS design permits publication of direct estimates for 22 large states and for the nation. Small sample sizes preclude publication of direct estimates for more detailed domains.  Past work on small area estimation with the NCVS data has focused heavily on geographic subdivisions of the United States. Estimates for demographic subdivisions are also of interest. We consider the problem of producing estimates for small domains defined by the intersection of age categories and genders. We construct estimates for four types of violent crimes in each of two time periods. We accomplish this through the use of a multivariate Bayesian hierarchical model. We compare a model with a log transformation to a model fit to the data in the original scale. We compare small area predictors based on a selected model to the direct estimators.