Calibration of Probabilities: The State of the Art

Abstract

An important criterion for evaluating probability assessors is their degree of calibration. A probability assessor is well calibrated if, over the long run, for all statements assigned a given probability the proportion that is true is equal to the probability assigned. For example, if you are well calibrated, then across all the many occasions that you assign a probability of .8, in the long run 80% of them should turn out to be true. If, instead, only 70% are true, you are not well calibrated, you are overconfident. If 95% of them are true, you are underconfident.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Aug 01, 1976
Accession Number
ADA033248

Entities

People

  • Baruch Fischhoff
  • L. D. Phillips
  • Sarah Lichtenstein

Tags

Communities of Interest

  • Biomedical
  • Human Systems

DTIC Thesaurus Topics

  • Accuracy
  • Applied Psychology
  • Business Administration
  • Calibration
  • Computer Programs
  • Detection
  • Engineering
  • Geography
  • Human Factors Engineering
  • Personnel Management
  • Probability
  • Probability Distributions
  • Psychology
  • Social Psychology
  • Social Sciences
  • Students
  • Systems Engineering

Readers

  • Computational Modeling and Simulation
  • Regression Analysis.