Prediction of Roadway Crashes Using Logistic Regression in SAS

International Journal of Computer Science and Engineering
© 2020 by SSRG - IJCSE Journal
Volume 7 Issue 10
Year of Publication : 2020
Authors : Srinivasan Suresh

pdf
How to Cite?

Srinivasan Suresh, "Prediction of Roadway Crashes Using Logistic Regression in SAS," SSRG International Journal of Computer Science and Engineering , vol. 7,  no. 10, pp. 13-17, 2020. Crossref, https://doi.org/10.14445/23488387/IJCSE-V7I10P103

Abstract:

Roadway crashes occur instantly with less time to respond. Predicting these crashes or identifying the major factors affecting these crashes can help to reduce these from occurring. As machine learning techniques help make these predictions and identify the impact factors, they can be applied to the roadway crash data set. The data set is obtained for the State of Virginia from the Department of Transportation. The logistic regression method was applied by grouping the dataset into fatal and non-fatal crashes. The model was built in SAS studio software and had an accuracy of 76%. The major factors were identified as Road, not lighted, Ramps, and Intersections on Divided roadways.

Keywords:

Fatal roadway crashes, Machine Learning, Logistic Regression, State of Virginia

References:

[1] VDOT. (2018, June 6). Crash Analysis Tool. https://public.tableau.com/profile/publish/Crashtools8_2/Main#!/publish-confirm
[2] National Center for Biotechnology Information (NCBI). (2009, Nov 06). Correspondence analysis is a useful tool to uncover the relationships among categorical variables. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3718710/ [3] Annapoorani Anantharaman, "A Study of Logistic Regression And Its Optimization Techniques Using Octave" SSRG International Journal of Computer Science and Engineering 6.10 (2019): 23-28.
[4] Virginia Roads (2018, March 15). Crash Data (Full Details). http://www.virginiaroads.org/datasets/crash-data-full-details
[5] Analytics Vidhya (2015, July 28). Beginners Guide to Learn Dimension Reduction Techniques
[6] https://www.analyticsvidhya.com/blog/2015/07/dimension-reduction-methods/
[7] Paul D. Allison (2008). Convergence failures in Logistic Regression. https://pdfs.semanticscholar.org/4f17/1322108dff719da6aa0d354d5f73c9c474de.pdf
[8] Virginia Department of Motor Vehicles. (, 2019). https://www.dmv.virginia.gov/safety/#crash_data/index.asp [9] Kononen, D. W., Flannagan C. A. & Wang S. C. (2011 Jan). Identification and validation of a logistic regression model for predicting serious injuries associated with motor vehicle crashes. Accid Anal Prev.;43(1):112-22. doi: 10.1016/j.aap.2010.07.018. [10] Feng, Z. X., Lu, S. S., Zhang, W. H., & Zhang, N. N. (2014). Combined prediction model of the death toll for road traffic accidents based on independent and dependent variables. Computational intelligence and neuroscience, 2014, 103196. https://doi.org/10.1155/2014/103196 [11] Dong C, Xie K, Sun X, Lyu M & Yue H. (2019) Roadway traffic crash prediction using a state-space model-based support vector regression approach. PLOS ONE 14(4): e0214866. https://doi.org/10.1371/journal.pone.0214866
[12] T.Maris Murugan, Dr.M.Kandasamy and G.Revathy,(2017). “Accident Prevention System through Intelligent Transport System”. SSRG International Journal of Industrial Engineering 4(1), 22-25.
[13] Annapoorani Anantharaman,(2019). “A Study of Logistic Regression And Its Optimization Techniques Using Octave”. SSRG International Journal of Computer Science and Engineering 6(10), 23-28.
[14] Sangeethu Sharma and Santini (2017). “Accident Avoidance and Safety System for Vehicular Communication”. SSRG International Journal of Industrial Engineering 4(2), 1-4.