|
Activity Number:
|
6
|
|
Type:
|
Invited
|
|
Date/Time:
|
Sunday, August 6, 2006 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Environmental and Ecological Statistics
|
| Abstract - #304953 |
|
Title:
|
Crime Mapping and Hotspot Detection
|
|
Author(s):
|
Reza Modarres*+ and Ganapati P. Patil
|
|
Companies:
|
The George Washington University and The Pennsylvania State University
|
|
Address:
|
Department of Statistics, Washington, DC, 20052,
|
|
Keywords:
|
hotspots ; geoinformatic surveillance ; scan statistic ; upper level set scan ; crime mapping
|
|
Abstract:
|
We discuss the Upper Level Scan (ULS) statistic and its theory as well as implementation for univariate and bivariate data. The ULS-tree and ULS-hotspot algorithm that depict and maintain a list of connected components of the rate surface at each level of the ULS tree are examined. The general theory of bivariate hotspot detection is discussed, including the bivariate binomial and Poisson models and the multivariate exceedance approach. We also examine the joint and intersection methods for detecting bivariate hotspots and study the sensitivity of the joint hotspots to the degree of association between the variables. We investigate the applications of ULS detection methodology in mapping of crime hotspots in the counties of several U.S. states.
|
