Dr. Regnier teaches practical decision science, including statistics, simulation, and mathematical programming, and how to use these tools effectively to make better decisions. Her research focuses on decisions under uncertainty when uncertainty is resolved over time. In particular, her work addresses how to formulate the decision problem and model the evolution of uncertainty so that they work together and produce good decisions. Dr. Regnier’s work has been funded by the National Science Foundation, Office of Naval Research, the Joint Typhoon Warning Center, the Marine Forces Reserve, and other Department of Defense organizations.
For Marines stationed in New Orleans, Dr. Regnier and Dr. Cameron MacKenzie built a simulator to give key personnel the opportunity to experience many synthetic storms to gain a familiarity with the evolution of Gulf hurricanes, forecast quality as a function of lead time, and their critical decision points. The simulator is still in development but has already been used in staff training exercises.
Details: MacKenzie, C.A. & Regnier, E. (2015). A Hurricane Decision Simulator for the U.S. Marine Corps Reserve Forces in New Orleans. Proceedings of the 44th Annual Meeting of the Western Decision Sciences Institute.
Probability forecasts are generated and consumed in many contexts. Examples include the National Hurricane Center’s wind-speed probability forecasts, statistical elections forecasts, and prediction market forecasts for geopolitical and other events. Often, these generate a time series of probability forecasts for each event. Dr. Regnier’s recent work on the properties required of efficient forecasting systems also provides statistical tests to diagnose inefficiency. Her ongoing work focuses on methods for improving these forecasts, for example, the 2014 US Senate election forecasts.
Details: Regnier, E. (2017). Probability forecasts made at multiple lead times. Management Science. Articles in Advance. DOI: 10.1287/mnsc.2016.2720.
Details: Selected Publications and Theses