Transaction Search Form: please type in any of the fields below.
Date: November 25, 2024 Mon
Time: 8:08 pm
Time: 8:08 pm
Results for crime forecasting
5 results foundAuthor: Berk, Richard Title: Asymmetric Loss Functions for Forecasting in Criminal Justice Settings Summary: The statistical procedures typically used for forecasting in criminal justice settings rest on symmetric loss functions. For quantitative response variables, overestimates are treated the same as underestimates. For categorical response variables, it does not matter in which class a case is inaccurately placed. In many criminal justice settings, symmetric costs are not responsive to the needs of stakeholders. It can follow that the forecasts are not responsive either. In this paper, the author considers asymmetric loss functions that can lead to forecasting procedures far more sensitive to the real consequences of forecasting errors. Theoretical points are illustrated with examples using criminal justice data of the kind that might be used for "predictive policing." Details: Philadelphia: University of Pennsylvania, Department of Statistics & Department of Criminology, 2010. 26p. Source: Working Paper Year: 2010 Country: United States URL: Shelf Number: 118398 Keywords: Crime Forecasting |
Author: Berk, Richard Title: The Role of Race in Forecasts of Violent Crime Summary: This paper addresses the role of forecasts of failure on probation or parole. Failure is defined as committing a homicide or attempted homicide or being the victim of a homicide or an attempted homicide. These are very rare events in the population of individuals studied, which can make these outcomes extremely difficult to forecast accurately. Building in the relative costs of false positives and false negatives, machine learning procedures are applied to construct useful forecasts. The central question addressed is what role race should play as a predictor when as an empirical matter the majority of perpetrators and victims are young, African-American, males. Details: Philadelphia: Department of Statistics, University of Pennsylvania, 2009. 29p. Source: Working Paper Year: 2009 Country: United States URL: Shelf Number: 118396 Keywords: Crime ForecastingHomicidesRaceViolent Crime |
Author: Harrell, Kim Title: The Predictive Accuracy of Hotspot Mapping of Robbery over Time and Space Summary: Police forces use hotspot mapping to provide a targeted approach to resource allocation, ensuring police officers are despatched to areas of high crime where their presence will have the most impact. Hotspot intelligence products are reliant on crime data sourced from police databases, and positional errors in this data will have an impact on the accuracy of the hotspot maps produced. The location of crime hotspots varies across both space and time. Despite this the use of temporal information is still rare because of the difficulties in pinpointing crime to an exact point in time, though crimes involving attended property provide the opportunity to record time more accurately. This research aimed to evaluate both the impact positional errors and the addition of temporal information have on the predictive accuracy of hotspot mapping of crime that inherently occurs in outdoor or public places through the utilisation of robbery data. Using robbery data recorded during a 24 month period (1st April 2011 - 31st March 2013) in the West Midlands Police Local Policing Unit of Birmingham South, the number and magnitude of positional errors present in the raw data was measured based on the Euclidean distance between recorded and actual locations of robbery offences. Positional errors ranging between 1 - 3766 metres were responsible for the suppression of a number of high intensity hotspots in the study area, and only 31% of all robberies had been allocated to the correct geographical location. To determine the influence of temporal information a mid-point measurement date was employed, and the ability of the retrospective robbery hotspots to predict the location of prospective robbery events measured, based on police shift periods, days of the week and spatial data alone. The results suggest that shift periods provide the best prospect for police forces utilising temporal information to improve the predictive ability of hotspot maps. Care needs to be taken to select a large enough dataset that will ensure sufficient clustering of crime points, and further research could be extended to incorporate different crime types. Details: Manchester, UK: University of Salford, Manchester, 2014. 159p. Source: Internet Resource: Dissertation: Accessed November 14, 2015 Year: 2014 Country: United Kingdom URL: Shelf Number: 137769 Keywords: Crime AnalysisCrime ClustersCrime ForecastingCrime HotspotsCrime MappingRobbery |
Author: Taylor, Ralph B. Title: Intra-Metropolitan Crime Patterning and Prediction Summary: This work examines the patterning and predictability of jurisdictional-level reported crime in the Philadelphia-Camden primary metropolitan statistical area. Spatial, temporal, and spatiotemporal features of the crime patterning are each investigated. The patterns are examined through three complementary lenses: the ecology of crime, the geography of crime, and the political economy of crime. Then, given what those inspections suggest, we examine the extent to which crime shifts prove predictable if we forecast a year ahead, or three years ahead. Details: Final report to the U.S. National Institute of Justice, 2015. 419p. Source: Internet Resource: Accessed April 6, 2016 at: https://www.ncjrs.gov/pdffiles1/nij/grants/249739.pdf Year: 2015 Country: United States URL: https://www.ncjrs.gov/pdffiles1/nij/grants/249739.pdf Shelf Number: 138576 Keywords: Crime ForecastingCrime Patterns Geography of Crime Spatial Analysis |
Author: Mohler, George O. Title: Randomized controlled field trials of predictive policing Summary: The concentration of police resources in stable crime hotspots has proven effective in reducing crime, but the extent to which police can disrupt dynamically changing crime hotspots is unknown. Police must be able to anticipate the future location of dynamic hotspots to disrupt them. Here we report results of two randomized controlled trials of near real-time Epidemic Type Aftershock Sequence (ETAS) crime forecasting, one trial within three divisions of the Los Angeles Police Department and the other trial within two divisions of the Kent Police Department (UK). We investigate the extent to which i) ETAS models of short term crime risk outperform existing best practice of hotspot maps produced by dedicated crime analysts, ii) police officers in the field can dynamically patrol predicted hotspots given limited resources, and iii) crime can be reduced by predictive policing algorithms under realistic law enforcement resource constraints. While previous hotspot policing experiments fix treatment and control hotspots throughout the experimental period, we use a novel experimental design to allow treatment and control hotspots to change dynamically over the course of the experiment. Our results show that ETAS models predict 1.4-2.2 times as much crime compared to a dedicated crime analyst using existing criminal intelligence and hotspot mapping practice. Police patrols using ETAS forecasts led to a average 7.4% reduction in crime volume as a function of patrol time, whereas patrols based upon analyst predictions showed no significant effect. Dynamic police patrol in response to ETAS crime forecasts can disrupt opportunities for crime and lead to real crime reductions. Details: Unpublished paper, 2015. 30p. Source: Internet Resource: Accessed September 23, 2016 at: http://paleo.sscnet.ucla.edu/MohlerEtAl-2015-JASA-Predictive-InPress.pdf Year: 2015 Country: United States URL: http://paleo.sscnet.ucla.edu/MohlerEtAl-2015-JASA-Predictive-InPress.pdf Shelf Number: 146052 Keywords: Crime ForecastingCrime MappingCriminal IntelligenceHot SpotsPredictive Policing |