Mathematics Colloquia and Seminars

Return to Colloquia & Seminar listing

Election dynamics are a rich complex system, and forecasting the upcoming U.S. elections is an exciting, high-stakes problem with many sources of subjectivity and uncertainty. In this talk, we take a dynamical-systems perspective on election forecasting, with the goal of helping to shed light on the forecast process and raise questions for future work. By adapting a well-studied model from epidemiology (a susceptible-infected-susceptible model), we show how to combine a compartmental approach with polling data to produce forecasts of presidential, senatorial, and gubernatorial elections at the state level. Our results for the last 16 years of U.S. elections are largely in agreement with those of popular analysts, and we apply our model to forecast the upcoming U.S. elections on 3 November 2020. We also use our modeling framework to explore how different methods for handling polling data and accounting for uncertainty affect forecasts. This is joint work with Samuel Chian, William He, Christopher Lee, Daniel Linder, Mason Porter, and Grzegorz Rempala.

Seminars this quarter will be online on Zoom. Please see the math bio seminar series email list or contact the organizers for link and password.