Higher Order Imprecise Probabilities and Statistical Testing

We generalize standard credal set models for imprecise probabilities to include higher order credal sets – confidences about confidences. In doing so, we specify how an agent’s higher order confidences (credal sets) update upon observing an event. Our model begins to address standard issues with imprecise probability models, like Dilation and Belief Inertia. We conjecture that when higher order credal sets contain all possible probability functions, then in the limiting case the highest order confidences converge to form a uniform distribution over the first order credal set, where we define uniformity in terms of the statistical distance metric (total variation distance). Finite simulation supports the conjecture. We further suggest that this convergence presents the total-variation-uniform distribution as a natural, privileged prior for statistical hypothesis testing.

[1]  Xiao-Li Meng,et al.  Judicious Judgment Meets Unsettling Updating: Dilation, Sure Loss and Simpson’s Paradox , 2017, Statistical Science.

[2]  Radoslav Harman,et al.  A conditional distribution approach to uniform sampling on spheres and balls in Lp spaces , 2012 .

[3]  A. Dempster.  Upper and lower probability inferences based on a sample from a finite univariate population. , 1967, Biometrika.

[4]  P. Walley,et al.  Imprecise Probabilities , 2000 .

[5]  A. Shemyakin.  Hellinger Distance and Non-informative Priors , 2014 .

[6]  Alan Hájek,et al.  What Conditional Probability Could Not Be , 2003, Synthese.

[7]  Lotfi A. Zadeh,et al.  Review of A Mathematical Theory of Evidence , 1984 .

[8]  Thomas Watson,et al.  Communication Complexity of Statistical Distance , 2018, Electron. Colloquium Comput. Complex..

[9]  Susanna Rinard.  Against Radical Credal Imprecision , 2013 .

[10]  C. Allen,et al.  Stanford Encyclopedia of Philosophy , 2011 .

[11]  Giuseppe Carlo Calafiore,et al.  Uniform sample generation in l/sub p/ balls for probabilistic robustness analysis , 1998, Proceedings of the 37th IEEE Conference on Decision and Control (Cat. No.98CH36171).

[12]  Larry Wasserman,et al.  Dilation for Sets of Probabilities , 1993 .

[13]  Kyburg,et al.  Randomness and the Right Reference Class , 1977 .

[14]  Peter Walley,et al.  Towards a unified theory of imprecise probability , 2000, Int. J. Approx. Reason..

[15]  James M. Joyce.  A DEFENSE OF IMPRECISE CREDENCES IN INFERENCE AND DECISION MAKING1 , 2010 .

[16]  Nathan P. Lemoine,et al.  Moving beyond noninformative priors: why and how to choose weakly informative priors in Bayesian analyses , 2019, Oikos.

[17]  E. Jaynes.  On the rationale of maximum-entropy methods , 1982, Proceedings of the IEEE.

[18]  P. Walley.  Statistical Reasoning with Imprecise Probabilities , 1990 .

[19]  P. Walley,et al.  A survey of concepts of independence for imprecise probabilities , 2000 .

[20]  Henry E. Kyburg,et al.  Bayesian and Non-Bayesian Evidential Updating , 1987, Artificial Intelligence.

[21]  E. Jaynes.  Information Theory and Statistical Mechanics , 1957 .

[22]  Glenn Shafer,et al.  A Mathematical Theory of Evidence , 2020, A Mathematical Theory of Evidence.

[23]  I. Levi.  On Indeterminate Probabilities , 1974 .

[24]  E. Wagenmakers.  A practical solution to the pervasive problems ofp values , 2007, Psychonomic bulletin & review.

[25]  Uninformative priors for Bayes’ theorem , 2002 .

[26]  Aron Vallinder,et al.  Imprecise Bayesianism and Global Belief Inertia , 2018, The British Journal for the Philosophy of Science.

[27]  An elementary proof that the Borel class of the reals has cardinality continuum , 2019, Acta Mathematica Hungarica.

[28]  L. M. M.-T..  Theory of Probability , 1929, Nature.