Summary Bayes factors are an increasingly popular tool for indexing evidence from experiments. For two competing population models, the factor reflects relative likelihood of observing some data under one model compared to other. Computing can be difficult, requiring integrate product and a prior distribution on parameter(s) both models. Previous work has obviated this difficulty independent-gr...

The decision sciences are increasingly challenged to advance methods for modeling analysts, accounting for both analytic strengths and weaknesses, to improve inferences taken from increasingly large and complex sources of data. We examine whether psychometric measures-personality, cognitive style, motivated cognition-predict analytic performance and whether psychometric measures are competitive...

We prove that if V = L then there is a Π11 maximal orthogonal (i.e. mutually singular) set of measures on Cantor space. This provides a natural counterpoint to the well-known Theorem of Preiss and Rataj [16] that no analytic set of measures can be maximal orthogonal.

For a large family of weights ρ in the unit disc and for fixed 1 < q < p < ∞, we give a characterization of those measures μ such that, for all functions f holomorphic in the unit disc, ‖f‖Lq(μ) ≤ C(μ) (∫ D |(1− |z|)f ′(z)|pρ(z) m(dz) (1− |z|)2 + |f(0)| ) 1 p .

We propose a class of nonparametric two-sample tests with a cost linear in the sample size. Two tests are given, both based on an ensemble of distances between analytic functions representing each of the distributions. The first test uses smoothed empirical characteristic functions to represent the distributions, the second uses distribution embeddings in a reproducing kernel Hilbert space. Ana...