Patrick T. Brandt

Patrick T. Brandt is a Professor of Political Science, Public Policy, and Political Economy at the University of Texas at Dallas, affiliated with the School of Economic, Political and Policy Sciences. His work integrates advanced statistical methods with political science, focusing on time series analysis, machine learning, and Bayesian modeling to study political dynamics.

His research spans international relations, political economy, terrorist targeting, and conflict forecasting. He specializes in developing novel models for event count time series, including the Bayesian Poisson Vector Autoregression and MS-BVAR packages for R. His NSF-funded projects focus on event data generation and real-time conflict forecasting.

Recent publications emphasize domain-specific language models (ConfliBERT variants), graph neural networks for conflict prediction, and machine translation challenges in political text analysis. He maintains the OpenEvent Data Repository and develops software like MSBVAR and PESTS for academic use.

Scientific Awards:

• Robert H. Durr Award for Best Methodology Paper, Midwest Political Science Association (2006)