Issues in the Analysis and Design of the Experimental Portion of the Health Insurance Study
Discusses two problems in the analysis of data and four problems in the design of the Health Insurance Study. Analytical problems discussed are measurement of price elasticity when price changes with total expenditure and measurement of the purchase of supplemental insurance. Design problems discussed are number of individuals to be assigned to any one plan, method of assignment, choice of individuals from given communities, and choice of communities. Includes brief description of Finite Selection Model, developed for this experiment to choose the optimal subset of observations (families) from a finite set of possible observations (families) in order to estimate a specified equation or equations.