Dr. Aaron Heuser

Dr. Aaron Heuser is a Senior Research Associate at IMPAQ International, working in the Data Science and Artificial Intelligence (DSAI) team. In this capacity, he serves as a lead mathematician, statistician, and computer scientist, in various projects involving modeling and simulation analysis.


Dr. Heuser’s work involves methodology spanning both frequentist and Bayesian approaches, encompassing machine learning, artificial intelligence, agent-based modeling, system dynamics, network analysis, and particle swarm optimization. In addition, Dr. Heuser possesses extensive experience implementing mathematical methods using coding languages such as Python and Java and platforms such as AWS-ML and Microsoft Azure.


Dr. Heuser has 10 years of experience working with a wide range of clients including the Food and Drug Administration, the U.S. Department of Labor, the Centers for Medicare and Medicaid Services, the National Institute of Justice, the Social Security Administration, the Newark Investment Board, and the National Institutes of Health (NIH).

Prior Experience

Prior to joining IMPAQ, Dr. Heuser served as a mathematical statistician with the NIH. During the first five months of his tenure, Dr. Heuser developed a new statistical test that helped his team win the Clinical Research Center Director’s Award for Science. Dr. Heuser aided in all aspects of research, from design to implementation. Research at the NIH was conducted with a team of intramural scientists, where Dr. Heuser specialized in the development of new mathematical and statistical methods, as well as coding and implementation. This work included the development of multistage compartmental models and agent-based models describing disability populations moving through different stages of disability, and the development of machine learning and text analytics models of structured, semi-structured, and unstructured data.


Dr. Heuser earned his Ph.D. from the University of Oregon in mathematics, with a focus on pure probability theory and stochastic analysis, in 2010. He has published and/or presented more than 10 papers on his work.

Featured IMPAQ Publications and Presentations

Heuser, A., Huynh, M., and Zhou, C. (2016, February). Introduction to Adaptive Designs [Conference workshop]. American Statistical Association Conference on Statistical Practices, San Diego.

IMPAQ Publications and Presentations

Kingi, H., Wang, L.A.D., Shafer, T., Huynh, M., Trinh, M., Heuser, A., Rochester, G., & Paredes, A. (2020). A numerical evaluation of the accuracy of influence maximization algorithms. Social Network Analysis and Mining, 10(1), 1–10.

Heuser, A., Huynh, M. & Chang, J.C. (2018). Asymptotic convergence in distribution of the area bounded by prevalence-weighted Kaplan–Meier curves using empirical process modelling. Royal Society Open Science, 5.

Heuser, A., Huynh, M., Zhang C., Kingi, H., Rochester, G., & Paredes, A. (2018). A system dynamics model for tobacco research. Proceedings of American Statistical Association Joint Statistical Meetings.

Heuser, A. & Huynh, M. (2017). An integrated framework for agent-based models, static and dynamic network analysis, and system dynamics [Presentation]. HHS/FDA Innovation Summit 2017, Office of the Chief Information Officer, Silver Spring, MD.

Chamberlain, A., Heuser, A., Selzer, A.K., Magill, K., Matite, M., Poe-Yamagata, E., & Barker, L.T. (2017). Evaluating the accessibility of American Job Centers for people with disabilities. Washington, DC: U.S. Department of Labor.

Previous Publications and Presentations

Rasch, E.K., Huynh, M., Ho, P.S., Heuser, A., Houtenville, A. & Chan, L. (2014). First in line: prioritizing receipt of social security disability benefits based on likelihood of death during adjudication. Medical Care, 52(11), 944–950.

Heuser, A. (2012). Generalized self intersection local time for a superprocess over a stochastic flow. Annals of Probability, 40(4), 1483–1534.

Heuser, A., Rasch, E., Huynh, M., Ho, P.S., & Chan, L. (2011, November). Analysis of the compassionate allowance (CAL) program: A systematic data-driven approach to identifying potential CAL conditions [Conference presentation]. 139th APHA Annual Meeting and Exposition.