Dr. Aaron Heuser

Dr. Aaron Heuser is a Senior Research Associate in the Advanced Analytics Division at IMPAQ International, LLC. In this capacity, Dr. Heuser 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, evolutionary algorithms, and particle swarm optimization. In addition, Dr. Heuser possesses extensive experience implementing mathematical and machine learning methods using coding languages such as Python and Java, and platforms such as AWS-ML and Azure Databricks.


Dr. Heuser has 10 years of experience participating in, supporting, and leading projects involving both quantitative and qualitative program evaluations and policy analyses for the public sector. His expertise spans a rich collection of methods, encompassing knowledge in both pure and applied mathematics and statistics, and varying data sources, including survey data, administrative data, electronic health records, and multi-dimensional image data. Dr. Heuser's diverse project work has provided him with a wide range of experiences and knowledge of all aspects of research from design to implementation. Clients include the Food and Drug Administration (FDA), Centers for Medicare and Medicaid Services (CMS), Department of Labor (DOL), National Institutes of Health (NIH), and Social Security Administration (SSA).

Prior Experience

Prior to joining IMPAQ, Dr. Heuser served as a mathematical statistician with the National Institutes of Health. 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 multi-stage 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 (focus in probability theory) in 2010.

IMPAQ Papers and Presentations

Huynh, M., Heuser, A. & Zhou, Chunxiao. (2016, February). Introduction to Adaptive Designs. Presentation at the American Statistical Association (ASA) Conference on Statistical Practice (CSP), San Diego, CA.

Previous Publications and Presentations

Heuser, A., Rasch, E., Huynh, M., Ho, P.S., Houtenville, A., & Chan, L. (2014, November). First in line: Prioritizing receipt of Social Security disability benefits based on likelihood of death during adjudication. Medical Care.

Heuser, A. (2012, July). Generalized Self-Intersection Local Time for a Superprocess Over a Stochastic Flow. Annals of Probability.

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 Paper: 139st APHA Annual Meeting and Exposition.