Proper local scoring rules on discrete sample spaces
arXiv:1104.2224 · doi:10.1214/12-AOS972
Abstract
A scoring rule is a loss function measuring the quality of a quoted probability distribution for a random variable , in the light of the realized outcome of ; it is proper if the expected score, under any distribution for , is minimized by quoting . Using the fact that any differentiable proper scoring rule on a finite sample space is the gradient of a concave homogeneous function, we consider when such a rule can be local in the sense of depending only on the probabilities quoted for points in a nominated neighborhood of . Under mild conditions, we characterize such a proper local scoring rule in terms of a collection of homogeneous functions on the cliques of an undirected graph on the space . A useful property of such rules is that the quoted distribution need only be known up to a scale factor. Examples of the use of such scoring rules include Besag's pseudo-likelihood and Hyvärinen's method of ratio matching.
arXiv commentPublished in at http://dx.doi.org/10.1214/12-AOS972 the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)
References in corpus (2)
Cited by in corpus (9)
- Theory and Applications of Proper Scoring Rules
- Proper local scoring rules
- Local proper scoring rules of order two
- Bayesian model comparison with the Hyvärinen score: computation and consistency
- Empirical AUC for evaluating probabilistic forecasts
- Maximum Entropy and Sufficiency
- Divergence and Sufficiency for Convex Optimization
- A Note on Bayesian Model Selection for Discrete Data Using Proper Scoring Rules
- Rejoinder to "Bayesian Model Selection Based on Proper Scoring Rules"