Epistemic Uncertainties: Dealing with a Lack of Knowledge
This chapter attempts to assess the impact of uncertainties that do not have a known distribution. In Sect. 12.1 we introduce a metric for evaluating the agreement between simulation and experiments. We then extend our focus to the case of agreement between experiment and simulation when there are calibration parameters in the simulation. We treat these cases using horsetail plots and p-boxes. To make a prediction on a new experiment, we can use the p-box to adjust a prediction, as shown in Sect. 12.4. Section 12.5 introduces Dempster-Shafer evidence theory for including expert judgment to construct p-boxes, followed by Sect. 12.6 where Kolomogorov-Smirnov confidence bounds are presented and Sect. 12.7 on Cauchy deviates.
- Agnesi M (1748) Instituzioni analitiche ad uso della gioventú italiana. Nella Regia-Ducal CorteGoogle Scholar
- Ferson S, Kreinovich V, Ginzburg L, Myers D, Sentz K (2003) Constructing probability boxes and dempster-shafer structures. Tech. Rep. SAND2002-4015, Sandia National LaboratoriesGoogle Scholar
- Ferson S, Kreinovich V, Hajagos J, Oberkampf W, Ginzburg L (2007) Experimental uncertainty estimation and statistics for data having interval uncertainty. Tech. Rep. SAND2007-0939, Sandia National LaboratoriesGoogle Scholar
- Kreinovich V, Nguyen HT (2009) Towards intuitive understanding of the Cauchy deviate method for processing interval and fuzzy uncertainty. In: Proceedings of the 2015 conference of the international fuzzy systems association and the european society for fuzzy logic and technology conference, pp 1264–1269Google Scholar
- Kreinovich V, Beck J, Ferregut C, Sanchez A, Keller G, Averill M, Starks S (2004) Monte-Carlo-type techniques for processing interval uncertainty, and their engineering applications. In: Proceedings of the workshop on reliable engineering computing, pp 15–17Google Scholar