Weighing Evidence: The Design and Comparison of Probability Thought Experiments.
Abstract
The assessment of probability on the basis of evidence is viewed as a thought experiment that yields an expression of degree of belief. Theories of subjective probability are viewed as tools or languages for analyzing evidence and expressing degree of belief. This article focuses on two probability languages: the classical Bayesian language and the language of belief functions. We describe and compare the semantics (i.e., the meaning of the scale) and the syntax (i.e., the formal calculus) of these languages. We also analyze the designs of thought experiments afforded by the two languages and discuss their implications. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 01, 1983
- Accession Number
- ADA131475
Entities
People
- Amos Tversky
- Glenn Shafer
Organizations
- Stanford University