A general model validation and testing tool. (March 2020)

• Record Type: Journal Article
• Authors: Vanslette, Kevin; Tohme, Tony; Youcef-Toumi, Kamal
• Is Part Of: Reliability engineering & system safety. Volume 195(2020)
• Publication Date: 2020-03
• Subjects: Verification and validation -- Uncertainty quantification -- Bayesian model testing -- Bayesian probability theory -- Inference -- Data science; Reliability (Engineering) -- Periodicals; System safety -- Periodicals; Industrial safety -- Periodicals; Fiabilité -- Périodiques; Sécurité des systèmes -- Périodiques; Sécurité du travail -- Périodiques; 620.00452
• DOI: 10.1016/j.ress.2019.106684
• Languages: English
• ISSNs: 0951-8320

Abstract

Highlights: The field of model validation may be condensed into a single probabilistic framework. This framework can reproduce each type of validation metrics as a special case. Limits are found in which several known validation metrics are equivalent. Novel validation metrics may be invented and quantified in this framework. This framework generalizes the Bayesian model testing framework. Abstract: We construct and propose the "Bayesian Validation Metric" (BVM) as a general model validation and testing tool. We find the BVM to be capable of representing all of the standard validation metrics (square error, reliability, probability of agreement, frequentist, area, probability density comparison, statistical hypothesis testing, and Bayesian model testing) as special cases and find that it can be used to improve, generalize, or further quantify their uncertainties. Thus, the BVM allows us to assess the similarities and differences between existing validation metrics in a new light. The BVM has the capacity to allow users to invent and select models according to novel validation requirements. We formulate and test a few novel compound validation metrics that improve upon other validation metrics in the literature. Further, we construct the BVM Ratio for the purpose of quantifying model selection under user defined definitions of agreement in the presence or absence of uncertainty. This construction generalizes the Bayesian model testing framework.