Kernel density estimation and hotspot mapping

Abstract

Purpose – The purpose of this paper is to examine the effects of user-defined parameters settings (e.g. interpolation method, grid cell size, and bandwidth) on the predictive accuracy of crime hotspot maps produced from kernel density estimation (KDE). Design/methodology/approach – The influence of variations in parameter settings on prospective KDE maps is examined across two types of interpersonal violence (e.g. aggravated assault and robbery) and two types of property crime (e.g. commercial burglary and motor vehicle theft). Findings – Results show that interpolation method has a considerable effect on predictive accuracy, grid cell size has little to no effect, and bandwidth as some effect. Originality/value – The current study advances the knowledge and understanding of prospective hotspot crime mapping as it answers the calls by Chainey et al. (2008) and others to further investigate the methods used to predict crime.

An Assessment of Police Officers’ Perception of Hotspots: What Can Be Done to Improve Officer’s Situational Awareness?

“…Supervisors should periodically map police officers' perceptions and examine them by making comparisons with crime distribution across atoms. An alternative is to use some of the spatial statistics methods (e.g., Getis Ord Gi*), which may offer grater statistical robustness but require some technical skills and are challenged with the subjectivity of their parameters' selection [39][40][41].…”

Predictive police patrolling to target hotspots and cover response demand

“…The method has also been proven to have the best predictive properties (Chainey et al 2008). This method represents crime as a continuous surface (Hart and Zandbergen 2014). Grid cells are used in the analysis but the area surrounding the grid cell is also considered within a certain bandwidth of the cell centre.…”

Understanding site selection of illegal border crossings into a fenced protected area: a rational choice approach

“…This produces a 'hotspot' map showing the spatial variation in rhino density across the study area. A KDE-value of at least twice the standard deviation above the mean was classified as a rhino density hotspot (Hart and Zandbergen 2014). Next, we divided the reserve's border into points with a 10-m interval and calculated the distance from each of these points to the nearest rhino density hotspot.…”

“…A crucial element in the KDE is the bandwidth parameter selection, but there is no general consensus on how to set these (Hart and Zandbergen 2014). We used five different approaches: average distance to K nearest neighbors, where K is the square root of the number of observations (Devroye et al 2013), cross validated bandwidth selection (Diggle 2003), the reference bandwidth (Calenge 2006), the method described by Vanek (2016), and a bandwidth used by the reserve based on visual assessments.…”