An information-based model for subjective probability

Intuitive subjective probability is commonly assumed to represent an individual’s comparative probability judgments of the events in some algebra. Conditions under which these judgments can be represented with numerical probabilities have been stated in various axiom systems. Existing axiom systems treat comparative probability judgments as a primitive concept without formulating the basis upon which these judgments are made. This paper argues for a constructive model of intuitive probability based on the relevant information available to an individual. In this model, probabilty is derived from the individual’s subjective judgments of the extent to which available information supports the occurrence of an event. A specific model is suggested and it is illustrated with some examples.