BYU

Abstract by Michael Vaughn

Personal Infomation


Presenter's Name

Michael Vaughn

Degree Level

Masters

Abstract Infomation


Department

Statistics

Faculty Advisor

Robert Richardson

Title

Comparing Statistical and Machine Learning Methods for Inference on Car Accidents in Allegheny County, PA

Abstract

Car crashes have unfortunately become more common in modern society. As accident rates increase over time, it is important to understand why they happen and how to reverse the trend. Federal and local government agencies are responsible for the legislation and city planning that could impact the severity of car crashes. This analysis employs the use of logistic regression, elastic net regression, and an implementation of gradient boosting to make inference about what factors have an effect on whether or not a crash is considered severe.