When to Refrain from Using Likelihood Surface Methods for Geographic Offender Profiling: An Ex Ante Test of Assumptions

Abstract

Likelihood surface methods for geographic offender profiling rely on several assumptions regarding the underlying location choice mechanism of an offender. We propose an ex ante test for checking whether a given set of crime locations is compatible with two necessary assumptions: circular symmetry and distance decay. The proposed (SDD) test compares the observed inter point distances of a given series of crimes with a theoretical distribution function governed by these assumptions, using a Monte Carlo simulation procedure for approximating that distribution function. We apply the SDD test to data on serial burglary from both the UK and the Netherlands. In most cases, the assumption of an underlying symmetric distance decay function has to be rejected.

Publication
Journal of Investigative Psychology and Offender Profiling, 8(3):242-256