|
Activity Number:
|
590
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Thursday, August 6, 2009 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Statistics in Defense and National Security
|
| Abstract - #305422 |
|
Title:
|
Using Variable Screening and Logistic Regression to Determine the Impact of Friendly Activity on Improvised Explosive Device Incidents
|
|
Author(s):
|
Darryl Ahner*+
|
|
Companies:
|
U.S. Military Academy
|
|
Address:
|
646 Swift Road, West Point, NY, 10996,
|
|
Keywords:
|
logistic regression ; prediction ; statistical model ; data mining
|
|
Abstract:
|
The use of Improvised Explosive Devices (IEDs) as a weapon of choice has emerged as one of the defining characteristics of Operation Iraqi Freedom. Numerous analytic efforts have focused on attempting to reveal the location of IED emplacement to mitigate the impact of their explosion and to disrupt the efforts of those insurgent networks that choose to utilize this asymmetric threat. This effort addresses the relationships between the tactical patterns and behaviors of the Coalition Force (CF) and the Threat's selection of the time and place for an intended IED attack. This presentation addresses this issue by developing a methodology that allows consideration of a large set of factors for each area and time period, determines which factors have influence, and uses logistic regression to determine the extent of those influences within the context of a larger system of influences.
|
