Title

Prediction/Causality Tradeoffs and Data Size Issues in Transportation Modeling: The Example of Highway-Safety Analysis

Time

1:40pm - 3:00pm

Location

Remote instruction

Abstract

The analysis of transportation data is largely dominated by traditional statistical methods (standard regression-based approaches), advanced statistical methods (such as models that account for unobserved heterogeneity), and data-driven methods (machine learning, neural networks, and so on). In the analysis of highway safety data, these methods have been applied mostly using data from observed crashes, but this can create a problem in uncovering causality since individuals that are inherently riskier than the population as a whole may be over-represented in the data. In addition, when and where individuals choose to drive could affect data analyses that use real-time data since the population of observed drivers could change over time. This issue, the size of the data (which can often influence the analysis method), and the implementation target of the analysis imply that analysts must often tradeoff the predictive capability (dominated by data-driven methods) and the ability to uncover the underlying causal nature of crash-contributing factors (dominated by statistical and econometric methods). However, the selection of the data-analysis method is often made without full consideration of this tradeoff, even though there are potentially important implications for the development of safety countermeasures and policies. This talk provides a discussion of the issues involved in this tradeoff with regard to specific methodological alternatives, and presents researchers with a better understanding of the trade-offs often being inherently made in their analysis.

Biographical Sketch

Fred Mannering is currently the Associate Dean for Research in the College of Engineering and a Professor of Civil and Environmental Engineering (with a courtesy appointment in Economics) at the University of South Florida. He received his undergraduate degree from the University of Saskatchewan, masters from Purdue University, and doctorate from the Massachusetts Institute of Technology. He was previously a professor at Penn State, A professor and Department Chair at the University of Washington, and School Head and Chaired professor at Purdue University. His research interests are in the application of econometric and statistical methods to the analysis of highway safety, transportation economics, vehicle demand, travel behavior and a variety of other engineering-related problems. He has published extensively with over 150 journal articles and two books: Principles of Highway Engineering and Traffic 001Analysis (now in its seventh edition) and Statistical and Econometric Methods for Transportation Data Analysis (now in its third edition). His body of work has been cited over 13,000 in Scopus, over 10,000 times in the Web of Science Core Collection, and over 25,000 times in Google Scholar. Dr. Mannering is currently Editor-in-Chief (and founding Editor) of the Elsevier Science journal Analytic Methods in Accident Research and previous Editor-in-Chief (2003-2012) and current Distinguished Editorial Board Member of the Elsevier Science journal Transportation Research Part B – Methodological. He is also the Interim Executive Director of the Center for Urban Transportation Research (CUTR) at the University of South Florida and an Associate Director of the TOMNET University Transportation Center.

Full resume available at: https://sites.google.com/view/fredmannering
Wikipedia page: https://en.wikipedia.org/wiki/Fred_Mannering