Item Details

Print View

Crime Hot Spot Forecasting With Data From the Pittsburgh [Pennsylvania] Bureau of Police, 1990-1998 [electronic resource]

Wilpen Gorr, Andreas Olligschlaeger
Format
Computer Resource; Online
Published
Ann Arbor, Mich. Inter-university Consortium for Political and Social Research [distributor] 2015
Edition
2015-08-07
Series
ICPSR
ICPSR (Series)
Access Restriction
AVAILABLE. This study is freely available to the general public.
Abstract

This study used crime count data from the Pittsburgh, Pennsylvania, Bureau of Police offense reports and 911 computer-aided dispatch (CAD) calls to determine the best univariate forecast method for crime and to evaluate the value of leading indicator crime forecast models.

The researchers used the rolling-horizon experimental design, a design that maximizes the number of forecasts for a given time series at different times and under different conditions. Under this design, several forecast models are used to make alternative forecasts in parallel. For each forecast model included in an experiment, the researchers estimated models on training data, forecasted one month ahead to new data not previously seen by the model, and calculated and saved the forecast error. Then they added the observed value of the previously forecasted data point to the next month's training data, dropped the oldest historical data point, and forecasted the following month's data point. This process continued over a number of months.

A total of 15 statistical datasets and 3 geographic information systems (GIS) shapefiles resulted from this study.

The statistical datasets consist of

  • Univariate Forecast Data by Police Precinct (Dataset 1) with 3,240 cases
  • Output Data from the Univariate Forecasting Program: Sectors and Forecast Errors (Dataset 2) with 17,892 cases
  • Multivariate, Leading Indicator Forecast Data by Grid Cell (Dataset 3) with 5,940 cases
  • Output Data from the 911 Drug Calls Forecast Program (Dataset 4) with 5,112 cases
  • Output Data from the Part One Property Crimes Forecast Program (Dataset 5) with 5,112 cases
  • Output Data from the Part One Violent Crimes Forecast Program (Dataset 6) with 5,112 cases
  • Input Data for the Regression Forecast Program for 911 Drug Calls (Dataset 7) with 10,011 cases
  • Input Data for the Regression Forecast Program for Part One Property Crimes (Dataset 8) with 10,011 cases
  • Input Data for the Regression Forecast Program for Part One Violent Crimes (Dataset 9) with 10,011 cases
  • Output Data from Regression Forecast Program for 911 Drug Calls: Estimated Coefficients for Leading Indicator Models (Dataset 10) with 36 cases
  • Output Data from Regression Forecast Program for Part One Property Crimes: Estimated Coefficients for Leading Indicator Models (Dataset 11) with 36 cases
  • Output Data from Regression Forecast Program for Part One Violent Crimes: Estimated Coefficients for Leading Indicator Models (Dataset 12) with 36 cases
  • Output Data from Regression Forecast Program for 911 Drug Calls: Forecast Errors (Dataset 13) with 4,936 cases
  • Output Data from Regression Forecast Program for Part One Property Crimes: Forecast Errors (Dataset 14) with 4,936 cases
  • Output Data from Regression Forecast Program for Part One Violent Crimes: Forecast Errors (Dataset 15) with 4,936 cases.
  • The GIS Shapefiles (Dataset 16) are provided with the study in a single zip file: Included are polygon data for the 4,000 foot, square, uniform grid system used for much of the Pittsburgh crime data (grid400); polygon data for the 6 police precincts, alternatively called districts or zones, of Pittsburgh(policedist); and polygon data for the 3 major rivers in Pittsburgh the Allegheny, Monongahela, and Ohio (rivers).

Cf: http://doi.org/10.3886/ICPSR03469.v1
Contents
  • Univariate Forecast Data by Police Precinct
  • Output Data from the Univariate Forecasting Program: Sectors and Forecast Errors
  • Multivariate, Leading Indicator Forecast Data by Grid Cell
  • Output Data from the 911 Drug Calls Forecast Program
  • Output Data from the Part One Property Crimes Forecast Program
  • Output Data from the Part One Violent Crimes Forecast Program
  • Input Data for the Regression Forecast Program for 911 Drug Calls
  • Input Data for the Regression Forecast Program for Part One Property Crimes
  • Input Data for the Regression Forecast Program for Part One Violent Crimes
  • Output Data from Regression Forecast Program for 911 Drug Calls: Estimated Coefficients for Leading Indicator Models
  • Output Data from Regression Forecast Program for Part One Property Crimes: Estimated Coefficients for Leading Indicator Models
  • Output Data from Regression Forecast Program for Part One Violent Crimes: Estimated Coefficients for Leading Indicator Models
  • Output Data from Regression Forecast Program for 911 Drug Calls: Forecast Errors
  • Output Data from Regression Forecast Program for Part One Property Crimes: Forecast Errors
  • Output Data from Regression Forecast Program for Part One Violent Crimes: Forecast Errors
  • GIS Shapefiles
Description
Mode of access: Intranet.
Notes
Title from ICPSR DDI metadata of 2016-02-11.
Series Statement
ICPSR 3469
ICPSR (Series) 3469
Other Forms
Also available as downloadable files.
Copyright Not EvaluatedCopyright Not Evaluated
Technical Details
  • Staff View

    LEADER 07161cmm a2200697la 4500
    001 ICPSR03469
    003 MiAaI
    006 m f a u
    007 cr mn mmmmuuuu
    008 160211s2015 miu f a eng d
    035
      
      
    a| (MiAaI)ICPSR03469
    040
      
      
    a| MiAaI c| MiAaI
    245
    0
    0
    a| Crime Hot Spot Forecasting with Data from the Pittsburgh [Pennsylvania] Bureau of Police, 1990-1998 h| [electronic resource] c| Wilpen Gorr, Andreas Olligschlaeger
    250
      
      
    a| 2015-08-07
    260
      
      
    a| Ann Arbor, Mich. b| Inter-university Consortium for Political and Social Research [distributor] c| 2015
    490
      
      
    a| ICPSR v| 3469
    516
      
      
    a| Numeric
    538
      
      
    a| Mode of access: Intranet.
    500
      
      
    a| Title from ICPSR DDI metadata of 2016-02-11.
    536
      
      
    a| United States Department of Justice. Office of Justice Programs. National Institute of Justice c| 1998-IJ-CX-K005
    506
      
      
    a| AVAILABLE. This study is freely available to the general public.
    530
      
      
    a| Also available as downloadable files.
    522
      
      
    a| Pennsylvania
    522
      
      
    a| Pittsburgh
    522
      
      
    a| United States
    520
    3
      
    a| <p> This study used crime count data from the Pittsburgh, Pennsylvania, Bureau of Police offense reports and 911 computer-aided dispatch (CAD) calls to determine the best univariate forecast method for crime and to evaluate the value of leading indicator crime forecast models. </p> <p> The researchers used the rolling-horizon experimental design, a design that maximizes the number of forecasts for a given time series at different times and under different conditions. Under this design, several forecast models are used to make alternative forecasts in parallel. For each forecast model included in an experiment, the researchers estimated models on training data, forecasted one month ahead to new data not previously seen by the model, and calculated and saved the forecast error. Then they added the observed value of the previously forecasted data point to the next month's training data, dropped the oldest historical data point, and forecasted the following month's data point. This process continued over a number of months. </p> <p> A total of 15 statistical datasets and 3 geographic information systems (GIS) shapefiles resulted from this study. </p> <p> The statistical datasets consist of </p> <p> <list type="bulleted"> <itm>Univariate Forecast Data by Police Precinct (Dataset 1) with 3,240 cases</itm> <itm>Output Data from the Univariate Forecasting Program: Sectors and Forecast Errors (Dataset 2) with 17,892 cases</itm> <itm>Multivariate, Leading Indicator Forecast Data by Grid Cell (Dataset 3) with 5,940 cases</itm> <itm>Output Data from the 911 Drug Calls Forecast Program (Dataset 4) with 5,112 cases</itm> <itm>Output Data from the Part One Property Crimes Forecast Program (Dataset 5) with 5,112 cases</itm> <itm>Output Data from the Part One Violent Crimes Forecast Program (Dataset 6) with 5,112 cases</itm> <itm>Input Data for the Regression Forecast Program for 911 Drug Calls (Dataset 7) with 10,011 cases</itm> <itm>Input Data for the Regression Forecast Program for Part One Property Crimes (Dataset 8) with 10,011 cases</itm> <itm>Input Data for the Regression Forecast Program for Part One Violent Crimes (Dataset 9) with 10,011 cases </itm> <itm>Output Data from Regression Forecast Program for 911 Drug Calls: Estimated Coefficients for Leading Indicator Models (Dataset 10) with 36 cases </itm> <itm>Output Data from Regression Forecast Program for Part One Property Crimes: Estimated Coefficients for Leading Indicator Models (Dataset 11) with 36 cases </itm> <itm>Output Data from Regression Forecast Program for Part One Violent Crimes: Estimated Coefficients for Leading Indicator Models (Dataset 12) with 36 cases </itm> <itm>Output Data from Regression Forecast Program for 911 Drug Calls: Forecast Errors (Dataset 13) with 4,936 cases </itm> <itm>Output Data from Regression Forecast Program for Part One Property Crimes: Forecast Errors (Dataset 14) with 4,936 cases </itm> <itm>Output Data from Regression Forecast Program for Part One Violent Crimes: Forecast Errors (Dataset 15) with 4,936 cases. </itm> <itm>The GIS Shapefiles (Dataset 16) are provided with the study in a single zip file: Included are polygon data for the 4,000 foot, square, uniform grid system used for much of the Pittsburgh crime data (grid400); polygon data for the 6 police precincts, alternatively called districts or zones, of Pittsburgh(policedist); and polygon data for the 3 major rivers in Pittsburgh the Allegheny, Monongahela, and Ohio (rivers). </itm></list></p>Cf: http://doi.org/10.3886/ICPSR03469.v1
    505
      
      
    t| Univariate Forecast Data by Police Precinct
    505
      
      
    t| Output Data from the Univariate Forecasting Program: Sectors and Forecast Errors
    505
      
      
    t| Multivariate, Leading Indicator Forecast Data by Grid Cell
    505
      
      
    t| Output Data from the 911 Drug Calls Forecast Program
    505
      
      
    t| Output Data from the Part One Property Crimes Forecast Program
    505
      
      
    t| Output Data from the Part One Violent Crimes Forecast Program
    505
      
      
    t| Input Data for the Regression Forecast Program for 911 Drug Calls
    505
      
      
    t| Input Data for the Regression Forecast Program for Part One Property Crimes
    505
      
      
    t| Input Data for the Regression Forecast Program for Part One Violent Crimes
    505
      
      
    t| Output Data from Regression Forecast Program for 911 Drug Calls: Estimated Coefficients for Leading Indicator Models
    505
      
      
    t| Output Data from Regression Forecast Program for Part One Property Crimes: Estimated Coefficients for Leading Indicator Models
    505
      
      
    t| Output Data from Regression Forecast Program for Part One Violent Crimes: Estimated Coefficients for Leading Indicator Models
    505
      
      
    t| Output Data from Regression Forecast Program for 911 Drug Calls: Forecast Errors
    505
      
      
    t| Output Data from Regression Forecast Program for Part One Property Crimes: Forecast Errors
    505
      
      
    t| Output Data from Regression Forecast Program for Part One Violent Crimes: Forecast Errors
    505
      
      
    t| GIS Shapefiles
    567
      
      
    a| Crime counts as reported in offense reports and 911 computer-aided dispatch (CAD) call records from the Pittsburgh, Pennsylvania, Bureau of Police
    650
      
    7
    a| crime 2| icpsr
    650
      
    7
    a| crime mapping 2| icpsr
    650
      
    7
    a| crime patterns 2| icpsr
    650
      
    7
    a| forecasting models 2| icpsr
    650
      
    7
    a| geographic distribution 2| icpsr
    650
      
    7
    a| geographic information systems 2| icpsr
    650
      
    7
    a| mapping 2| icpsr
    650
      
    7
    a| police effectiveness 2| icpsr
    650
      
    7
    a| police records 2| icpsr
    650
      
    7
    a| prediction 2| icpsr
    650
      
    7
    a| trends 2| icpsr
    653
    0
      
    a| ICPSR XVII.E. Social Institutions and Behavior, Crime and the Criminal Justice System
    653
    0
      
    a| NACJD VII. Crime and Delinquency
    700
    2
      
    a| Gorr, Wilpen u| Carnegie Mellon University
    700
    2
      
    a| Olligschlaeger, Andreas u| Carnegie Mellon University
    710
    2
      
    a| Inter-university Consortium for Political and Social Research.
    830
      
    0
    a| ICPSR (Series) v| 3469
    856
    4
    0
    u| http://proxy.its.virginia.edu/login?url=http://doi.org/10.3886/ICPSR03469.v1
    999
      
      
    w| WEB l| INTERNET m| UVA-LIB t| INTERNET
▾See more
▴See less

Availability

Access Online