Martin, Jan Niklas and Uhrmacher, Adelinde M. (2026) Challenges in calibrating simulation models with stylized facts. In: 40th ACM SIGSIM Conference on Principles of Advanced Discrete Simulation, 24-26 June 2026, Vienna, Austria. Proceedings, published by Association for Computing Machinery (ACM), New York, NY, USA, pp. 193-194.
Full text not available from this repository.Abstract
Often, no suitable data are available for validating and calibrating simulation models. Stylized facts are empirical and domain-specific insights that can be used to express expected model behavior. They are already being widely used to validate simulation models in specific application domains. The use of stylized facts for the automatic calibration of simulation models raises several challenges: the design of a domain-specific language that permits a succinct, intuitive, and unambiguous representation of a broad range of stylized facts; the formulation of a quantitative metric measuring the satisfaction of stylized facts to steer the calibration; and the effective communication of both the language semantics and the results. We discuss first approaches to address these challenges.
| Item Type: | Conference or Workshop Item (Paper) |
|---|
