Converting qualitative assessments to quantitative assumptions: Bayes' rule and the pundit's wager

When developing quantitative scenarios, it is necessary to choose parameter values to fit a particular story line, such as growth rates and resource-use efficiencies. This paper describes one approach to making such choices using Bayesian statistical reasoning to systematically combine data from a reference set (e.g., historical data) with qualitative assessments regarding the scenario. The result is a probability distribution, rather than a single value, reflecting the non-unique correspondence between a qualitative description of a scenario and its quantitative realization. The paper also describes a simple software tool that implements the method.