Suchergebnisse

This article presents a generalization of the Condorcet Jury Theorem. All results to date assume a fixed value for the competence of jurors, or alternatively, a fixed probability distribution over the possible competences of jurors. In this article, we develop the idea that we can learn the competence of the jurors by the jury vote. We assume a uniform prior probability assignment over the competence parameter, and we adapt this assignment in the light of the jury vote. We then compute the posterior probability, conditional on the jury vote, of the hypothesis voted over. We thereby retain the central results of Condorcet, but we also show that the posterior probability depends on the size of the jury as well as on the absolute margin of the majority. [Reprinted by permission of Sage Publications Ltd., copyright holder.]

This article presents a generalization of the Condorcet Jury Theorem. All results to date assume a fixed value for the competence of jurors, or alternatively, a fixed probability distribution over the possible competences of jurors. In this article, we develop the idea that we can learn the competence of the jurors by the jury vote. We assume a uniform prior probability assignment over the competence parameter, and we adapt this assignment in the light of the jury vote. We then compute the posterior probability, conditional on the jury vote, of the hypothesis voted over. We thereby retain the central results of Condorcet, but we also show that the posterior probability depends on the size of the jury as well as on the absolute margin of the majority.

Abstract
This article comments on the article of Thorn and Schurz in this volume and focuses on, what we call, the problem of parasitic experts. We discuss that both meta- induction and crowd wisdom can be understood as pertaining to absolute reliability rather than comparative optimality, and we suggest that the involvement of reliability will provide a handle on this problem.

In this article, we introduce the concept of model uncertainty. We review the frequentist and Bayesian ideas underlying model selection, which serve as an introduction to the rest of this special issue on 'All models are wrong...', a workshop under the same name was held in March 2011 in Groningen to critically examined the field of statistical model selection methods over the past 40 years. We briefly introduce the philosophical debate that is concerned with model selection. We present the results of a questionnaire that was distributed under the participants of the workshop, showing that the field has not yet reached a comforting consensus and is still in full swing.