Forecasting residential burglary

Derek Deadman

doi:10.1016/s0169-2070(03)00091-8

Abstract

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.

Highlights

Thenum berofrecorded residentialburglariesin England and W aleshasbeen declining since1993.Such adeclineisneitheruniqueto thiscategory of crim enorjustlim ited to England and W ales,assim ilarpatternsin recorded crim eoverthisperiod havebeen experienced in m any otherEuropean countriesand in theU nited States(seeField (1999),p.16).W hilstthereare severalpublished econom etricanalysesofrecorded crim e,untilrecently only one(D eadm an and Pyle(1997))appearsto havebeen used for prediction.Itisthereforem ostencouraging thatasignificantbreak w ith past practicehasbeen m adew ith thepublication of H om eO fficeResearch Study No.198 (Dhirietal(1999)) addressing theissueofm odelling and predicting property crim etrends
TheH om eO fficepredictionshavebeen m adeon thebasisofforecastsof dem ographicand econom icchangesand on theassum ption thatno other factors(such asgovernm entpoliciesto reducecrim e)arealtered overthe prediction period.Itispredicted thattherecentdeclinesin recorded property crim ewillbereversed from 1999 onw ardsand thelevelofsuch crim ew ill risesubstantiallyby 2001.Thesearedram aticand bravepredictions,w hich are clearly ofdeep concern to politiciansm aking policyin thisareaand to Chief Constablesw ho m ay berequired to m eettargetssetforthem by policy m akers.Such predictionsarealso ofconsiderableinterestto thegeneral publicand they did receivew idespread presscoverageon release
W hatiscom m on to theeconom etricpredictionsin thispaperand those presented by theH om eO fficeisthatthey areboth obtained from m odels incorporating an error-correction term .In both casescrim eism odelled as having along run equilibrium solution togetherwith am echanism which allow sfor dynam icadjustm entto thislong run path from positionsoffthis path.Thisstructureappearsto beofcentralim portancein thepattern of predictionsdiscussed below,w hich w ereobtained from thefollow ing m odel: O rdinary LeastSquaresEstim ation D ependentV ariableis∆ Resburg

Summary

Introduction

Thenum berofrecorded residentialburglariesin England and W aleshasbeen declining since1993.Such adeclineisneitheruniqueto thiscategory of crim enorjustlim ited to England and W ales,assim ilarpatternsin recorded crim eoverthisperiod havebeen experienced in m any otherEuropean countriesand in theU nited States(seeField (1999),p.16).W hilstthereare severalpublished econom etricanalysesofrecorded crim e,untilrecently only one(D eadm an and Pyle(1997))appearsto havebeen used for prediction.Itisthereforem ostencouraging thatasignificantbreak w ith past practicehasbeen m adew ith thepublication of H om eO fficeResearch Study No.198 (Dhirietal(1999)) addressing theissueofm odelling and predicting property crim etrends. Theestim ated m odelused hereforprediction ofthenum berofrecorded residentialburglary offencesincludeseconom icactivity variables (consum ption and unem ploym ent),crim inaljusticevariables(probability of conviction,probability ofim prisonm ent,length ofsentenceand num berof police)and adem ographicvariable(num berof m alesaged 15-24).The m odelw asestim ated in naturallogarithm icform using annualdatafrom 1950 to 1997 and incorporated adum m y variableto take accountofachangein recording practicefollow ing theTheftA ctof1968. In contrast,theH om eO fficem odelpredictsseparately both theftand burglary using justtw o explanatory variables,nam ely thestock ofgoods (proxied in each yearby thesum oftotalhousehold finalconsum ption expenditurein thecurrentand threepreceding years)and thenum berof m alesaged 15 and 20.Thetheftand burglary seriesw ereadjusted to take accountofthenew counting rulesintroduced in theTheftA ct.A nnualdata for1951 to 1998 w ereused in theestim ation ofthem odelw hich w asalso in logarithm icform. W hatiscom m on to theeconom etricpredictionsin thispaperand those presented by theH om eO fficeisthatthey areboth obtained from m odels incorporating an error-correction term .In both casescrim eism odelled as having along run equilibrium solution togetherwith am echanism which allow sfor dynam icadjustm entto thislong run path from positionsoffthis path.Thisstructureappearsto beofcentralim portancein thepattern of predictionsdiscussed below ,w hich w ereobtained from thefollow ing m odel:

Dum m y Intercept

5.Conclusion

Findings

BIBLIO G RA PH Y