Using Geographically Weighted Regression Models to Analyze Crash Harm
Date of Award
Master of Science (MS)
Civil & Environmental Engineering
Asad J. Khattak
Call Number for Print
Special Collections LD4331.E542 Z54 2009
Society pays a high cost for collisions in terms of property damage, injuries and loss of life. This paper has two objectives. One is to examine various factors associated with harm caused by highway collisions. Both global OLS and local GWR models are provided to identify specific variables influencing crash harm. Another objective is to demonstrate the use of a Geographically Weighted Regression (GWR) method in this complex safety problem, which can estimate models for each crash location and show spatial variation. The results show that the GWR model is a significant improvement on the OLS model; the test for non-stationarity shows the relationship between angle collision, fixed object off road collision, road width and percentage of truck traffic and crash harm change significantly over space; four maps of local estimator of these four variables and detailed analysis are provided to exhibit the considerable spatial variation. The results provide valuable information on high-risk factors associated with collision harm and the spatial characteristics for these associations.
In Copyright. URI: http://rightsstatements.org/vocab/InC/1.0/ This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).
"Using Geographically Weighted Regression Models to Analyze Crash Harm"
(2009). Master of Science (MS), Thesis, Civil & Environmental Engineering, Old Dominion University, DOI: 10.25777/rx79-8685