The Subjectivist View of Decision-Making.

Abstract

The subjectivist, Bayesian paradigm for a decision-maker is described. It is shown how the notion of utility, and the principle of maximizing expected utility both depend on the description of uncertainty through probability. The justification for the necessity of this description due to de Finetti is outlined. The twin, practical problems of the evaluation of the decision-maker's probabilities and utilities are discussed. Probability, as used in the paradigm, is a subjectivist notion which is distinct from the chance, or frequentist, concept and there is discussion of this difference. The calculations for the analysis of a decision tree are described and the notions of the utility of data developed. The statistical analysis of data that flows from the paradigm is described and the basic, likelihood principle derived and discussed. The material is illustrated by a simple example from insurance. (Author)

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jul 01, 1981
Accession Number
ADA103879

Entities

People

  • Dennis V. Lindley

Organizations

  • University of Wisconsin–Madison

Tags

Communities of Interest

  • Human Systems

DTIC Thesaurus Topics

  • Accidents
  • Aircrafts
  • Bayes Theorem
  • Dynamic Programming
  • Information Science
  • Insurance
  • Mathematics
  • New York
  • North America
  • Operations Research
  • Probability
  • Probability Distributions
  • Risk
  • Statistical Analysis
  • Statistics
  • Test And Evaluation
  • United States

Readers

  • Artificial Intelligence
  • Statistical inference.
  • Systems Analysis and Design

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Machine Learning Algorithms