Description:

Abstract:

This talk discusses five data analytic issues that are often misunderstood or overlooked completely in data analysis: unobserved heterogeneity, data size and prediction/causality trade-offs, selectivity/identification, temporally shifting model parameters, and compensating user behaviors. The nuances of these issues are discussed in the context of contemporary data-analytic methods including traditional statistical methods (standard regression-based approaches), advanced statistical methods (such as models that account for unobserved heterogeneity), data-driven methods (machine learning, neural networks, and so on) and causal-inference models (commonly used in the economics literature). In transportation contexts, commonly used data can be afflicted with one or more of the aforementioned issues. For example, in the safety field, analytic methods have been applied mostly using data from observed crashes, but this can create a selectivity/identification problem because individuals that are inherently riskier may be over-represented in the data, creating a selectivity/identification problem (a problem can also afflict non-traditional data such as those extracted from video analytics and instrumented vehicles). This selectivity/identification problem, and additional complications resulting from unobserved heterogeneity, temporal behavioral shifts and possible compensating behaviors, often require that analysts consider the size of the data (which can often influence the selection of analysis method), and the tradeoff between predictive capability (dominated by data-driven methods) and the ability to uncover the underlying causal nature the process being studied (dominated by statistical and econometric methods). Regrettably, 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 effective transportation policies. This talk concludes with a discussion of the various elements involved in this tradeoff with regard to specific methodological alternatives.

Bio: 

Dr. Fred Mannering is the Executive Director of the Center for Urban Transportation Research and a Professor in the Department of Civil and Environmental Engineering at the University of South Florida. He was previously the Charles Pankow Professor of Civil Engineering at Purdue University, Head of Civil Engineering at Purdue, Professor and Chair of Civil and Environmental Engineering at the University of Washington and an Assistant Professor at the Pennsylvania State University. He received his undergraduate degree from the University of Saskatchewan, masters from Purdue University, and PhD from the Massachusetts Institute of Technology. His body of work has been cited over 30,000 times in Google Scholar, over 13,000 times in the Web-of-Science core collection database, and over 17,000 times in Scopus. Dr. Mannering has published over 160 refereed journal articles, two widely adopted textbooks, over sixty other publications (conference proceedings, project reports, book reviews and commentaries), and has given over 150 invited lectures, keynote speeches, and presentations at professional conferences. 
He is Founding Editor and currently Editor-in-Chief of Elsevier Science’s Analytic Methods in Accident Research (2021 Web-of-Science Journal Impact Factor 14.556) and was previously Editor-in-Chief of Elsevier Science’s Transportation Research Part B: Methodological. Some of his awards and recognitions include: Web-of-Science Highly Cited Researcher (2019, 2020, 2021); Council of University Transportation Centers Lifetime Achievement Award (2021); most highly-cited author (highest total citations and citations per paper) in the 50-year history of the journal Accident Analysis and Prevention (2020); Eno Foundation’s “Top 10 Transportation Thought Leaders in Academia” (2016); Charles B. Murphy Outstanding Undergraduate Teaching Award (Purdue University's highest undergraduate teaching honor, 2013); induction into Purdue University’s “Book of Great Teachers” (2013); and the Arthur M. Wellington Prize (2010), James Laurie Prize (2009), and Wilbur S. Smith Award (2005) all awarded by the American Society of Civil Engineers.