Crime Forecasting Research Articles

Exploring data assimilation and forecasting issues for an urban crime model

In this paper, we explore some of the various issues that may occur in attempting to fit a dynamical systems (either agent- or continuum-based) model of urban crime to data on just the attack times and locations. We show how one may carry out a regression analysis for the model described by Shortet al.(2008,Math. Mod. Meth. Appl. Sci.) by using simulated attack data from the agent-based model. It is discussed how one can incorporate the attack data into the partial differential equations for the expected attractiveness to burgle and the criminal density to predict crime rates between attacks. Using this predicted crime rate, we derive a likelihood function that one can maximise in order to fit parameters and/or initial conditions for the model. We focus on carrying out data assimilation for two different parameter regions, namely in the case where stationary and non-stationary crime hotspots form. It is found that the likelihood function is ‘flat’ for large ranges of parameters, and that this has major implications for crime forecasting. Hence, we look at how one might carry out a goodness-of-fit and forecasting analysis for crime rates given the range of parameter fits. We show how one can use the Kolmogorov–Smirnov statistic to assess the goodness-of-fit. The dynamical systems analysis of the partial differential equations proves invaluable to understanding how the crime rate forecasts depend on the parameters and their sensitivity. Finally, we outline several interesting directions for future research in this area where we believe that the combination of dynamical systems modelling, analysis, and data assimilation can prove effective in developing policing strategies for urban crime.

Randomized Controlled Field Trials of Predictive Policing

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 (United Kingdom). 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 an 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.

How do offenders choose where to offend? Perspectives from animal foraging

PurposeResearch suggests that offender spatial decision‐making is not random. However, little is known about if or how offences in a series influence where an offender will target next. Drawing on concepts and empirical findings from environmental criminology and the ecology literature, in this article I consider what spatial patterns might be expected in the sequential crimes committed by serial offenders and provide an empirical example.MethodsData for detected burglars are analysed and patterns in the inter‐event distances for sequential offences compared with those signatures typically associated with three types of foraging behaviour – central place foraging, Brownian walks and Lévy walks. Analyses involve the use of a Monte Carlo simulation to derive an expected distribution for central place foraging, while the observed probability density function of sequential inter‐event distances is compared to exponential and power law distributions to test for evidence of Brownian and Lévy walks, respectively.ResultsAnalyses suggest that patterns in burglar sequential inter‐event distances cannot be explained by a simple central place foraging strategy. The distribution of sequential inter‐event distances is found to be consistent with both Brownian and Lévy walks.ConclusionsThe findings suggest that there are regularities in the sequential spatial choices made by offenders, and that these are similar to those observed across species. Reasons for why there is evidence of both Brownian and Lévy walks are discussed. The implications of the findings for forensic techniques such as crime linkage analysis, geographic offender profiling and crime forecasting are discussed.

Short-term Crime Forecasting in Kedah

Abstract Crime Forecasting is rarely used globally by police including Malaysia. In practice usually the police would target persons with their criminality and studying their strategy of implementing crime. The police will also monitor the current crime situation and will take necessary action when the crime index increases. Both of these scenarios require action taken after crime incurred. Therefore if crime forecasting can be adopted perhaps early crime prevention can be enforced. The aim of this study is to identify crime patterns in Kedah using univariate forecasting technique. Seventy six recent monthly data (January 2006 – April 2012) were obtained from IPK Alor Star with the permission from PDRM Bukit Aman. Exploratory Data Analysis (EDA) and adjusted decomposition technique were conducted in order to fulfil the objective of the study. The findings revealed that total crimes in Kedah were mainly contributed by type of property crime (80-85%) while violent crime has a small proportion only. Fortunately due to the productiveness of the police the property crime trend indicated curve declining pattern. As for the violent crime, the index was very low and displayed straight forward line (no trend). Significant seasonal pattern were identified between property crime and with all major festive seasons in Malaysia (i.e. Chinese New Year, Eid-ulFitr, Eid-ulAdha and Deepavali) together with all school breaks. Therefore the seasonal index can be a good indicator to be adopted in crime fighting strategy by the police.

Hybrid Support Vector Regression and Autoregressive Integrated Moving Average Models Improved by Particle Swarm Optimization for Property Crime Rates Forecasting with Economic Indicators

Crimes forecasting is an important area in the field of criminology. Linear models, such as regression and econometric models, are commonly applied in crime forecasting. However, in real crimes data, it is common that the data consists of both linear and nonlinear components. A single model may not be sufficient to identify all the characteristics of the data. The purpose of this study is to introduce a hybrid model that combines support vector regression (SVR) and autoregressive integrated moving average (ARIMA) to be applied in crime rates forecasting. SVR is very robust with small training data and high-dimensional problem. Meanwhile, ARIMA has the ability to model several types of time series. However, the accuracy of the SVR model depends on values of its parameters, while ARIMA is not robust to be applied to small data sets. Therefore, to overcome this problem, particle swarm optimization is used to estimate the parameters of the SVR and ARIMA models. The proposed hybrid model is used to forecast the property crime rates of the United State based on economic indicators. The experimental results show that the proposed hybrid model is able to produce more accurate forecasting results as compared to the individual models.

Joint Utility of Event-Dependent and Environmental Crime Analysis Techniques for Violent Crime Forecasting

Violent crime incidents occurring in Irvington, New Jersey, in 2007 and 2008 are used to assess the joint analytical capabilities of point pattern analysis, hotspot mapping, near-repeat analysis, and risk terrain modeling. One approach to crime analysis suggests that the best way to predict future crime occurrence is to use past behavior, such as actual incidents or collections of incidents, as indicators of future behavior. An alternative approach is to consider the environment in which crimes occur and identify features of the landscape that would be conducive to crime. Thanks to advances in geographic information system technology and federally funded (free) software applications such as CrimeStat III or the Near Repeat Calculator, these methods have recently been made more accessible to “average” users. This study explores the information products that each method offers for the purposes of place-based violent crime forecasting and resource allocation. Findings help to answer questions about where, when, and why violent crimes occur in a jurisdiction. Ways in which event-dependent and environmental crime analysis techniques can be utilized as complementary instruments in a crime analyst’s tool kit are discussed in detail.

Crime Forecasting and Planning in Developing Countries: Emerging Issues

In recent times, there is the growing manifestation among stake holders that crime cannot be controlled exclusively through the action of the police and criminal justice administrators. This reality has brought about the idea of new forms of approach on crime prevention against the backdrop of the apparent failure of the police, courts and prisons to stem the rising crime rates in many societies. Thus, the prevalence of crime necessitates the study of geographical and communal setting of environment in order to adopt a most effective approach to tackling crime. The paper reiterates that criminal statistics is a tool for answering questions, helps to approach the study of crime and justice from the scientifi c world and the facts obtained may aid in dealing with the crime problem, but reliance on crime statistics must be done with circumspection. The study reveals that many crimes are committed but never reported to the police, while of those reported, many are not recorded, and of those recorded, and many are not summarized or reflected in statistical tables. The initial stages of the criminalization process depend heavily upon the victims commitment to “making a complaint” which entails their acceptance of the burdens that may follow, with a preparedness to take the matter all the way to court and, if necessary, to give evidence. Also, the victims’ attitudes a times affect the figures generated for crime statistics. In some other instances, the police may not record a crime reported, if the police authority in their own discretion regard such incident as purely domestic matter or those that are too trivial to be investigated. The paper concludes that crime is assuming a frightening dimension as virtually everybody is becoming vulnerable to criminal attacks, brutal acts of terrorism, armed robbery, assassination, ritual killings, cultism, bombing, kidnapping and the like that are now prevalent and very alarming. Key words: Crime; Police; Criminal Justice; Administrators; Statistics; Complaint; Terrorism; Armed robbery; Assassination; Killings; Bbombing; Kidnapping

A brief history of the analysis of crime concentration

Decades of research demonstrate that crime is concentrated at a range of spatial scales. Such findings have clear implications for crime forecasting and police resource allocation models. More recent work has also shown that crime clusters in spaceandtime with a regularity that might improve methods of crime prediction. In this paper I review some of the available evidence and provide illustrations of the types of analysis – spatial and spatio-temporal – conducted hitherto. With a few exceptions, the application of formal Mathematics in the study of space–time patterns of crime has been rather limited, and so a central aim of the paper is to stimulate interest in this area of research.

Offender as Forager? A Direct Test of the Boost Account of Victimization

Recent research has demonstrated that burglary clusters in space and time, resulting in temporal changes in crime hotspot patterns. Offender foraging behavior would yield the observed pattern. The offender as forager hypothesis is tested by analyzing patterns in two types of acquisitive crime, burglary and theft from motor vehicle (TFMV). Using a technique developed to detect disease contagion confirms that both crime types cluster in space and time as predicted, but that the space-time clustering of burglary is generally independent of that for TFMV. Police detections indicate that crimes of the same type occurring closest to each other in space and time are those most likely to be cleared to the same offender(s), as predicted. The implications of the findings for crime forecasting and crime linkage are discussed.

Stable and Fluid Hotspots of Crime: Differentiation and Identification

It has been known for some time that crime risk is unevenly distributed and that some geographic areas experience more crime than others. However, less attention has been given to the ebb and flow of crime hotspots and in particular the stability of crime problems. This is important as the identification of appropriate crime reduction responses should be informed by the timing and stability of crime problems as well as their location. This paper reviews recent work concerned with space-time patterns of crime and the implications of the findings for crime forecasting. Variation in the stability of crime hotspots is discussed in some detail and an empirical demonstration provided to illustrate the central issues. A modification of a current hotspotting method that may better identify stable hotspots of crime is also presented. Findings are discussed in relation to their policy implications including those that focus on the built environment.

Space–Time Patterns of Risk: A Cross National Assessment of Residential Burglary Victimization

Using epidemiological techniques for testing disease contagion, it has recently been found that in the wake of a residential burglary, the risk to nearby homes is tem- porarily elevated. This paper demonstrates the ubiquity of this phenomenon by analyzing space-time patterns of burglary in 10 areas, located in five different countries. While the precise patterns vary, for all areas, houses within 200 m of a burgled home were at an elevated risk of burglary for a period of at least two weeks. For three of the five countries, differences in these patterns may partly be explained by simple differences in target density. The findings inform theories of crime concentration and offender targeting strategies, and have implications for crime forecasting and crime reduction more generally.

Introduction to crime forecasting

This short paper introduces the six papers comprising the Special Section on Crime Forecasting. A longer title for the section could have been “Forecasting crime for policy and planning decisions and in support of tactical deployment of police resources.” Crime forecasting for police is relatively new. It has been made relevant by recent criminological theories, made possible by recent information technologies including geographic information systems (GIS), and made desirable because of innovative crime management practices. While focused primarily on the police component of the criminal justice system, the six papers provide a wide range of forecasting settings and models including UK and US jurisdictions, long- and short-term horizons, univariate and multivariate methods, and fixed boundary versus ad hoc spatial cluster areal units for the space and time series data. Furthermore, the papers include several innovations for forecast models, with many driven by unique features of the problem area and data.

Forecasting residential burglary

Following the work of Dhiri et al. [Modelling and predicting property crime trends. Home Office Research Study 198 (1999). London: HMSO] at the Home Office predicting recorded burglary and theft for England and Wales to the year 2001, econometric and time series models were constructed for predicting recorded residential burglary to the same date. A comparison between the Home Office econometric predictions and the less alarming econometric predictions made in this paper identified the differences as stemming from the particular set of variables used in the models. However, the Home Office and one of our econometric models adopted an error correction form which appeared to be the main reason why these models predicted increases in burglary. To identify the role of error correction in these models, time series models were built for the purpose of comparison, all of which predicted substantially lower numbers of residential burglaries. The years 1998–2001 appeared to offer an opportunity to test the utility of error correction models in the analysis of criminal behaviour. Subsequent to the forecasting exercise carried out in 1999, recorded outcomes have materialised, which point to the superiority of time series models compared to error correction models for the short-run forecasting of property crime. This result calls into question the concept of a long-run equilibrium relationship for crime.

Short-term forecasting of crime

The major question investigated is whether it is possible to accurately forecast selected crimes 1 month ahead in small areas, such as police precincts. In a case study of Pittsburgh, PA, we contrast the forecast accuracy of univariate time series models with naı̈ve methods commonly used by police. A major result, expected for the small-scale data of this problem, is that average crime count by precinct is the major determinant of forecast accuracy. A fixed-effects regression model of absolute percent forecast error shows that such counts need to be on the order of 30 or more to achieve accuracy of 20% absolute forecast error or less. A second major result is that practically any model-based forecasting approach is vastly more accurate than current police practices. Holt exponential smoothing with monthly seasonality estimated using city-wide data is the most accurate forecast model for precinct-level crime series.

Criminal incident prediction using a point-pattern-based density model

Law enforcement agencies need crime forecasts to support their tactical operations; namely, predicted crime locations for next week based on data from the previous week. Current practice simply assumes that spatial clusters of crimes or “hot spots” observed in the previous week will persist to the next week. This paper introduces a multivariate prediction model for hot spots that relates the features in an area to the predicted occurrence of crimes through the preference structure of criminals. We use a point-pattern-based transition density model for space–time event prediction that relies on criminal preference discovery as observed in the features chosen for past crimes. The resultant model outperforms the current practices, as demonstrated statistically by an application to breaking and entering incidents in Richmond, VA.