Situational and Victim Correlates of Increased Case Fatality Rates in Los Angeles Shootings, 2005-2021.

The gun assault case fatality rate measures the fraction of shooting victims who die from their wounds. Considerable debate has surrounded whether gun assault case fatality rates have changed over time and what factors may be involved. We use crime event data from Los Angeles to examine the victim and situational correlates of gun assault case fatality rates over time. We estimated log binomial regression models for the probability of death in each year from 2005 to 2021, conditioned on situational and victim characteristics of the crime. Case fatality rates increased by around 1.3% per year between 2005 and 2021 from around 15.9 to 19.7%. Baseline case fatality rates differed systematically by most situational and victim but followed similar temporal trends. Only victim age significantly covaried with the temporal trend in case fatality rates. An individual shot in Los Angeles in 2021 was 23.7% more likely to die than the equivalent victim in 2005. The steady increase in case fatality rates suggests that there were around 394 excess fatalities over what would have occurred if case fatality rates remained at the 2005 level. Increases in the average age of victims over time may contribute to the general temporal trend. We hypothesize that older victims are more likely to be shot indoors where lethal close-range wounds are more likely.

Are DNA data a valid source to study the spatial behavior of unknown offenders?

Studying the spatial behaviour of unknown offenders (i.e. undetected offenders) is difficult, because police recorded crime data do not contain information about these offenders. Recently, forensic DNA data has been used to study unknown offenders. However, DNA data are only a subset of the crimes committed by unknown offenders stored in police recorded crime data. To establish the suitability of DNA data for studying the spatial offending behaviour of unknown offenders, we examine the concentration and spatial similarity of detected but unsolved crimes in police recorded crime data (N = 181,483) and DNA data (N = 1913) over 27 Belgian judicial districts for four crime types. We established spatial similarity for certain crime types (in some districts). This offers opportunities for DNA data to be used to study unknown offenders' spatial offending behaviour. Implications for theory and research are discussed.

Spatio-temporal stochastic differential equations for crime incidence modeling.

We propose a methodology for the quantitative fitting and forecasting of real spatio-temporal crime data, based on stochastic differential equations. The analysis is focused on the city of Valencia, Spain, for which 90247 robberies and thefts with their latitude-longitude positions are available for a span of eleven years (2010-2020) from records of the 112-emergency phone. The incidents are placed in the 26 zip codes of the city (46001-46026), and monthly time series of crime are built for each of the zip codes. Their annual-trend components are modeled by Itô diffusion, with jointly correlated noises to account for district-level relations. In practice, this study may help simulate spatio-temporal situations and identify risky areas and periods from present and past data.

Making the most of cybercrime and fraud crime report data: a case study of UK Action Fraud.

Introduction: Researchers and public authorities are increasingly exploring the potential of administrative data to generate new insights. This includes recent work leveraging the opportunities of the crime report data collected by the UK's national reporting centre Action Fraud (AF). However, the quality of these data and its implications for data users have not been systematically analysed. Objectives: This paper outlines challenges and opportunities of using AF data in cybercrime and fraud victimisation research and practice and makes recommendations to improve the quality of this dataset. Methods: The author has undertaken two studies using samples of AF data pertaining to crime reports within the Welsh police forces, between 2014 and 2020. Quality diagnostic checks, reflections and methodological decisions were considered across each study. These were reviewed, key themes were identified and discussed with data users and a broader group of researchers to finalise the recommendations presented. Results: The strengths and limitations of AF data are discussed and grouped into themes, closely aligned with four quality dimensions widely used by statistical authorities. This includes an assessment of 1) the impact of under-reporting and 2) the purpose and rules of crime recording, on the relevance of the data to its users; 3) the accuracy and reliability of the data; 4) the consistency of recording and its impact on coherence and comparability; and 5) the accessibility and timeliness of the data. Conclusions: Recommendations are made to improve AF data to generate better quality insights across the dimensions of relevance, accuracy & reliability, coherence & comparability and the accessibility & timeliness of this dataset. Additionally, a data catalogue would enable frontline officers and researchers to make the most of this dataset, harnessing it to produce key insights for crime prevention, investigation, and victim support.

Combining Benford's Law and machine learning to detect money laundering. An actual Spanish court case.

OBJECTIVES: This paper is based on the analysis of the database of operations from a macro-case on money laundering orchestrated between a core company and a group of its suppliers, 26 of which had already been identified by the police as fraudulent companies. In the face of a well-founded suspicion that more companies have perpetrated criminal acts and in order to make better use of what are very limited police resources, we aim to construct a tool to detect money laundering criminals. METHODS: We combine Benford's Law and machine learning algorithms (logistic regression, decision trees, neural networks, and random forests) to find patterns of money laundering criminals in the context of a real Spanish court case. RESULTS: After mapping each supplier's set of accounting data into a 21-dimensional space using Benford's Law and applying machine learning algorithms, additional companies that could merit further scrutiny are flagged up. CONCLUSIONS: A new tool to detect money laundering criminals is proposed in this paper. The tool is tested in the context of a real case.