Estimation and Uncertainty Quantification of Stochastic Systems
Abstract
Analysts in every field face the challenge of how to best use available data to estimate performance, quantify uncertainty, and predict the future. The data is almost never Òjust right,Ó but rather scarce, excessive, corrupted, uncertain, and incomplete. External information derived from experiences, established Òlaws,Ó and physical restrictions offer opportunities to remedy the situation and should be utilized. Applications in sustainable energy, military operations, natural resources, image reconstruction, uncertainty quantification, and reliability engineering are rich with problems where decisions rely on data analysis under such circumstances. We address these problems within a framework that identifies a function that according to some criterion best represents the given data set and satisfies constraints derived from the data as well as external information. Epi-splines provide the linchpin that allows us to handle shape restrictions, information growth, and approximations.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jun 25, 2021
- Source ID
- W911NF1210273
Entities
People
- Roger J-B Wets
Organizations
- Army Contracting Command
- United States Army
- University of California, Davis