Profile
Patrick T. Brandt (Ph. D. Indiana University, M.S. Northwestern University, A.B. College of William and Mary) studies political methodology, research methods, political economy, and interstate and intrastate political conflict and violence.
His recent work has focused on a National Science Foundation sponsored project to forecast conflict and violence, particularly recent conflicts in the Balkans, Middle East, the Indian subcontinent and Asia.
Other related research looks at the linkages between national economies and government approval in the U.S. and elsewhere. Most of this work applies Bayesian time series analysis to forecast and understand political events.
He recently published (with John T. Williams) a book on multivariate time series models.
He teaches courses in research methods, research design, decision theory and American politics.
Past Work Experience
- 2001-2005
- Assistant Professor, Department of Political Science, University of North Texas
- 2000-2001
- Visiting Lecturer of American Politics and Methodology, Department of Political Science, Indiana University, Bloomington
- January-June 2000
- Research Fellow, Harvard-MIT Data Center
Awards
“Bayesian Time Series Models for the Analysis of International Conflict” (joint with John R. Freeman). National Science Foundation. 2004-2006.
Professional Organizations
- American Political Science Association
- Midwest Political Science Association
- Society for Political Methodology
Courses
- GOVT 2302: Political Institutions in the U.S. and Texas
- GOVT 4305: Introduction to Research Methodology
- PSCI 5362: Multivariate Models
- PSCI 6325: Decision Theory
Media Expertise
Forecasting political conflict and violence, forecasting elections, political changes over time.
Publications
Brandt, Patrick T. and John T. Williams. (2007). Modeling Multiple Time Series. Beverly Hills: Sage.
Brandt, Patrick T. and John Freeman (2006). “Advances in Bayesian Time Series Modeling and the Study of Politics: Theory Testing, Forecasting, Policy Analysis.” Political Analysis, 14(1): 1-36.
Brandt, Patrick T. (2002) “Using the Right Tools for Time Series Data Analysis” The Political Methodologist 10(2): 22–27.
Brandt, Patrick T. and John T. Willaims (2001). “A Linear Poisson Autoregressive Model: The Poisson AR(p)” Political Analysis. 9(2): 164-184.
Brandt, Patrick T., John T. Williams, Benjamin O. Fordham, and Brian Pollins (2000). “Dynamic Modeling for Persistent Event Count Time Series” American Journal of Political Science. 44(4); 823-843.
Additional working papers are available at http://www.utdallas.edu/~pbrandt/
Software
“PESTS: Poisson Estimators for State-space Time Series.'' with John T. Williams. 1998. Gauss and R programs for estimating event count time series models. GAUSS and R versions at http://www.utdallas.edu/~pbrandt/codepage.html
“MSBVAR: Bayesian estimators and inferences for vector autoregression (VAR) models'', an R package for VAR and BVAR models. R package, documentation, working papers and examples at http://yule.utdallas.edu
- Updated: October 18, 2006
