Test

, Volume 5, Issue 1, pp 1–60

Scoring rules and the evaluation of probabilities

  • R. L. Winkler
  • Javier Muñoz
  • José L. Cervera
  • José M. Bernardo
  • Gail Blattenberger
  • Joseph B. Kadane
  • Dennis V. Lindley
  • Allan H. Murphy
  • Robert M Oliver
  • David Ríos-Insua
Article
  • 498 Downloads

Summary

In Bayesian inference and decision analysis, inferences and predictions are inherently probabilistic in nature. Scoring rules, which involve the computation of a score based on probability forecasts and what actually occurs, can be used to evaluate probabilities and to provide appropriate incentives for “good” probabilities. This paper review scoring rules and some related measures for evaluating probabilities, including decompositions of scoring rules and attributes of “goodness” of probabilites, comparability of scores, and the design of scoring rules for specific inferential and decision-making problems

Keywords

Attributes of “Good” probabilities Decomposition of expected Scores Evaluation of Probabilities Probability Assessment Probability Forecasts Scoring Rules 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© SEIO 1996

Authors and Affiliations

  • R. L. Winkler
    • 1
  • Javier Muñoz
    • 2
  • José L. Cervera
    • 3
  • José M. Bernardo
    • 4
  • Gail Blattenberger
    • 5
  • Joseph B. Kadane
    • 6
  • Dennis V. Lindley
    • 7
  • Allan H. Murphy
    • 8
  • Robert M Oliver
    • 9
  • David Ríos-Insua
    • 10
  1. 1.Fuqua School of Business and Institute of Statistics and Decision SciencesDuke UniversityDurhamUSA
  2. 2.Generalitat ValencianaSpain
  3. 3.Instituto Nacional de EstadísticaSpain
  4. 4.Universitat de ValènciaSpain
  5. 5.University of UtahUSA
  6. 6.Carnegie Mellon UniversityUSA
  7. 7.MineheadUK
  8. 8.Prediction and Evaluation SystemsUSA
  9. 9.University of California at BerkeleyUSA
  10. 10.Universidad Politécnica de MadridMadridEspana

Personalised recommendations