This is the last of 10 chapters on "Spatial Modeling II" from the user manual for CrimeStat IV, a spatial statistics package that can analyze crime incident location data.
This chapter, "The CrimeStat Time Series Forecasting Module," provides guidance on the features and use of the CrimeStat Time Series Forecasting module, which is designed to forecast crime counts or counts of any type of events, as well as the early detection of unusual levels of activity in current data. A single run detects and forecasts all districts that compose a jurisdiction. The single interface page requires the user to specify an input file, either the primary file or another file, identify variables in the file used for forecasting, select a seasonality adjustment, specify an exponential smoothing model, turn on Trigg tracking signals, use default values or choose Trigg parameter values, and save the output. An example of the running of CrimeStat's Time Series Forecasting module uses Pittsburgh's monthly crime data for December 1999. The chapter advises that a major limitation of any approach to working with crime time series data for the tactical deployment of police resources is that the size of area units must be small, patrol district and smaller, but then the associated time series data has relatively low crime counts, and any estimated models have sizable estimation and prediction errors. In such a situation, it is better to use simple models, if for no other reason than there is little else to get out of the limited data than what simple models can find. The chapter recommends using the automated detection method presented for early warning of crime increases in conjunction with crime mapping and other sources of information to diagnose and respond to emerging crime-area problems. 7 figures, including computer screens and maps from examples, and 7 references
Downloads
Related Datasets
Similar Publications
- Place-based Correlates of Motor Vehicle Theft and Recovery Measuring Spatial Influence Across Neighbourhood Context
- Integrating Population Health Data on Violence Into the Emergency Department: A Feasibility and Implementation Study
- Crime Mode Split Modeling (CrimeStat IV: A Spatial Statistics Program for the Analysis of Crime Incident Locations, Version 4.0)