Statistical Inference for Multi-Objective Simulation Optimization with Application to Health Policy Analysis

Interdisciplinary Areas: Data and Engineering Applications

Project Description

The research objective of the postdoctoral fellowship is to make fundamental advances in statistical inference for multi-objective simulation optimization (MOSO) with health policy analysis as the principal application context. Modeling and simulation is an increasingly important tool for the design, analysis, and optimization of modern complex systems, including public health systems. When evaluating potential public health interventions, policymakers often need to supplement clinical studies with computational simulation-based studies. This practice leads to simulation-based cost-effectiveness optimization, which is a MOSO problem. However, the theoretical underpinnings required for statistical inference in a MOSO context are severely under-developed. The fellow will create theory, methods, and algorithms to 1) construct confidence regions, estimate variance, and estimate bias to quantify the risk and error associated with using a stochastic simulation model to estimate the Pareto set and trade-off rates of different polices; 2) learn the Pareto set optimally using simulation; and 3) quantify model uncertainty for calculating the effect of model errors on decision-making. The fellow will also implement the algorithms and design software that specifically guides public policy development for the prevention of health diseases and stroke through simulation-based cost-effectiveness optimization.


Start Date

July 1, 2024


Postdoc Qualifications

PhDs in Operations Research, Decision Analytics, Statistics, Applied Mathematics, Computer Science, or similar areas.

Candidates shall be well trained in probability theory and have experience with advanced methodological topics in stochastic optimization and simulation optimization.

Candidates who have conducted research in health economics or more specifically, in health policy simulation analysis are preferred.



Nan Kong, PhD. Professor, Weldon School of Biomedical Engineering.

Susan Hunger, PhD. Associate Professor, School of Industrial Engineering.

Raghu Pasupathy, PhD. Professor, Department of Statistics.


Short Bibliography

1. Li, Y., N. Kong, M. A. Lawley, L. Weiss, and J. A. Pagán (2015). “Advancing the Use of Evidence-Based Decision Making in Local Health Departments with Systems Science Methodologies.” American Journal of Public Health. Vol. 105, No. S2, pp. S217-S222.
2. Lou, Z., S. S. Yi, J. Pomeranz, R. Suss, R. Russo, P. E. Rummo, H. Eom, J. Liu, Y. Zhang, A. Moran, B. Bellows, N. Kong, and Y. Li (2023). “The Health and Economic Impact of Using a Sugar Sweetened Beverage Tax to Fund Fruit and Vegetable Subsidies in New York City: A Modeling Study.” Journal of Urban Health. Vol. 100, No. 1, pp. 51-62. DOI: 10.1007/s11524-022-00699-3.
3. Bress AP*, Bellows BK*, King JB, Hess R, Beddhu S, Zhang Z, Berlowitz DR, Conroy MB, Fine L, Oparil S, Morisky DE, Kazis LE, Ruiz-Negrón N, Powell J, Tamariz L, Whittle J, Wright J, Supiano MA, Cheung AK, Weintraub WS, Moran AE, for the SPRINT Research Group. Cost-effectiveness of intensive versus standard blood pressure control. N Engl J Med. 2017 Aug 24;377(8):745-755.
4. Hunter, S. R.; and Pasupathy, R. Central Limit Theorems for constructing confidence regions in strictly convex multi-objective simulation optimization. In Feng, B.; Pedrielli, G.; Peng, Y.; Shashaani, S.; Song, E.; Corlu, C. G.; Lee, L. H.; Chew, E. P.; Roeder, T.; and Lendermann, P, editor(s), Proceedings of the 2022 Winter Simulation Conference, pages 3015–3026, Piscataway, NJ, 2022. IEEE
5. Z. Su, R. Pasupathy, Y. Yeh, and P. W. Glynn. Overlapping Batch Confidence Intervals on Statistical Functionals Constructed from Time Series: Application to Quantiles, Optimization, and Estimation, under review, 2022.