Spatiotemporal Correlations in Criminal Offense Records

With the increased availability of rich behavioral datasets, we present a novel application of tools to analyze this information. Using criminal offense records as an example, we employ cross-correlation measures, eigenvalue spectrum analysis, and results from random matrix theory to identify spatiotemporal patterns on multiple scales. With these techniques, we show that most significant correlation exists on the time scale of weeks and identify clusters of neighborhoods whose crime rates are affected simultaneously by external forces.