2019/02/07 by Rafael Prieto-Curiel, Rafael Prieto Curiel, Curiel, Rafael Prieto · 2 citations
Engineering · Mathematics · Physics and Astronomy · Social Sciences · #Computer science #Crime Patterns and Interventions #Criminology #Data mining #Degree (music) #Econometrics #Economics #Engineering #FOS: Mathematics #FOS: Physical sciences #Mathematics #Measure (data warehouse) #Metric (unit) #Null (SQL) #Physics #Physics and Society (physics.soc-ph) #Probability (math.PR) #Randomness #Sensitivity (control systems) #Sociology #Statistics #math.PR #physics.soc-ph
paper · pdf · doi:10.48550/arxiv.1902.03105
published in arXiv (Cornell University) (Cornell University)
arxiv created 2019/02/07 · openalex publication_date 2019/02/07 · arxiv updated 2019/02/11 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Crime is highly concentrated in a few places, is committed by a few offenders and is suffered by a few victims. In recent decades, the concentration of crime has become an accepted fact, yet, little is known in terms of how to measure this concentration of crime such that the metric takes into account the fact that crime has, in general, a low frequency, it fluctuates, it is highly concentrated and has a certain degree of randomness. Here, the most frequently used metrics for concentration of crimes are reviewed. A null model with complete randomness is used for comparing between different concentration metrics, which allows constructing a sensitivity analysis for every metric against varying crime rates. Results show that most ways of measuring the concentration of crime are in fact, showing only that crime is rare or that it has fluctuations, but fail to work as a method to compare the concentration of crime between different regions, types of crime or across time.