Title: Spatio-temporal Bayesian modeling of crime in Philadelphia
Authors: Shane Jensen - The Wharton School of the University of Pennsylvania (United States) [presenting]
Abstract: Urban data analysis has been recently improved through publicly available high-resolution data, allowing us to investigate theories in criminology and urban design empirically. We will focus on a particular direction: spatial-temporal modeling of the change in crime over the past decade in the city of Philadelphia. We will explore different parametric and non-parametric Bayesian approaches for finding regions of the city that share similar crime dynamics. Within this context, we have developed a methodology for non-parametric clustering of regions simultaneously across multiple levels of spatial resolution. We will also provide an interpretation of our results in the context of the geography and built environment of the city of Philadelphia.