Search

In the literature on imprecise probability, little attention is paid to the fact that imprecise probabilities are precise on a set of events. We call these sets systems of precision . We show that, under mild assumptions, the system of precision of a lower and upper probability form a so-called (pre-)Dynkin system. Interestingly, there are several settings, ranging from machine learning on partial data over frequential probability theory to quantum probability theory and decision making under uncertainty, in which, a priori, the probabilities are only desired to be precise on a specific underlying set system. Here, (pre-)Dynkin systems have been adopted as systems of precision, too. We show that, under extendability conditions, those pre-Dynkin systems equipped with probabilities can be embedded into algebras of sets. Surprisingly, the extendability conditions elaborated in a strand of work in quantum probability are equivalent to coherence from the imprecise probability literature. On this basis, we spell out a lattice duality which relates systems of precision to credal sets of probabilities. We conclude the presentation with a generalization of the framework to expectation-type counterparts of imprecise probabilities. The analogue of pre-Dynkin systems turns out to be (sets of) linear subspaces in the space of bounded, real-valued functions. We introduce partial expectations, natural generalizations of probabilities defined on pre-Dynkin systems. Again, coherence and extendability are equivalent. A related but more general lattice duality preserves the relation between systems of precision and credal sets of probabilities.

Meta-analysis is a statistical method used in evidence synthesis for combining, analyzing and summarizing studies that have the same target endpoint and aims to derive a pooled quantitative estimate using fixed and random effects models or network models. Differences among included studies depend on variations in target populations (ie, heterogeneity) and variations in study quality due to study design and execution (ie, bias). The risk of bias is usually assessed qualitatively using critical appraisal, and quantitative bias analysis can be used to evaluate the influence of bias on the quantity of interest. We propose a way to consider ignorance or ambiguity in how to quantify bias terms in a bias analysis by characterizing bias with imprecision (as bounds on probability) and use robust Bayesian analysis to estimate the overall effect. Robust Bayesian analysis is here seen as Bayesian updating performed over a set of coherent probability distributions, where the set emerges from a set of bias terms. We show how the set of bias terms can be specified based on judgments on the relative magnitude of biases (ie, low, unclear, and high risk of bias) in one or several domains of the Cochrane's risk of bias table. For illustration, we apply a robust Bayesian bias-adjusted random effects model to an already published meta-analysis on the effect of Rituximab for rheumatoid arthritis from the Cochrane Database of Systematic Reviews., (© 2022 The Authors. Statistics in Medicine published by John Wiley & Sons Ltd.)

This paper describes a robust and computationally feasible method to train and quantify the uncertainty of Neural Networks. Specifically, we propose a back propagation algorithm for Neural Networks with interval predictions. In order to maintain numerical stability we propose minimising the maximum of the batch of errors at each step. Our approach can accommodate incertitude in the training data, and therefore adversarial examples from a commonly used attack model can be trivially accounted for. We present results on a test function example, and a more realistic engineering test case. The reliability of the predictions of these networks is guaranteed by the non-convex Scenario approach to chance constrained optimisation, which takes place following training, and is hence robust to the performance of the optimiser. A key result is that, by using minibatches of size M, the complexity of the proposed approach scales as O(M⋅N iter ), and does not depend upon the number of training data points as with other Interval Predictor Model methods. In addition, troublesome penalty function methods are avoided. To the authors' knowledge this contribution presents the first computationally feasible approach for dealing with convex set based epistemic uncertainty in huge datasets., (Copyright © 2019 Elsevier Ltd. All rights reserved.)

The contribution proposes to model imprecise and uncertain reasoning by a mental probability logic that is based on probability distributions. It shows how distributions are combined with logical operators and how distributions propagate in inference rules. It discusses a series of examples like the Linda task, the suppression task, Doherty's pseudodiagnosticity task, and some of the deductive reasoning tasks of Rips. It demonstrates how to update distributions by soft evidence and how to represent correlated risks. The probabilities inferred from different logical inference forms may be so similar that it will be impossible to distinguish them empirically in a psychological study. Second-order distributions allow to obtain the probability distribution of being coherent. The maximum probability of being coherent is a second-order criterion of rationality. Technically the contribution relies on beta distributions, copulas, vines, and stochastic simulation.

Measuring the accuracy of diagnostic tests is crucial in many application areas including medicine, machine learning and credit scoring. The receiver operating characteristic (ROC) curve is a useful tool to assess the ability of a diagnostic test to discriminate between two classes or groups. In practice, multiple diagnostic tests or biomarkers are combined to improve diagnostic accuracy. Often, biomarker measurements are undetectable either below or above the so-called limits of detection (LoD). In this paper, nonparametric predictive inference (NPI) for best linear combination of two or more biomarkers subject to limits of detection is presented. NPI is a frequentist statistical method that is explicitly aimed at using few modelling assumptions, enabled through the use of lower and upper probabilities to quantify uncertainty. The NPI lower and upper bounds for the ROC curve subject to limits of detection are derived, where the objective function to maximize is the area under the ROC curve. In addition, the paper discusses the effect of restriction on the linear combination's coefficients on the analysis. Examples are provided to illustrate the proposed method. Copyright © 2017 John Wiley & Sons, Ltd., (Copyright © 2017 John Wiley & Sons, Ltd.)