Optimal design of lottery with cumulative prospect theory

A lottery is a popular form of gambling between a seller and multiple buyers, and its profitable design is of primary interest to the seller. Designing a lottery requires modeling the buyer decision-making process for uncertain outcomes. One of the most promising descriptive models of such decision-making is the cumulative prospect theory (CPT), which represents people's different attitudes towards gain and loss, and their overestimation of extreme events. In this study, we design a lottery that maximizes the seller's profit when the buyers follow CPT. The derived problem is nonconvex and constrained, and hence, it is challenging to directly characterize its optimal solution. We overcome this difficulty by reformulating the problem as a three-level optimization problem. The reformulation enables us to characterize the optimal solution. Based on this characterization, we propose an algorithm that computes the optimal lottery in linear time with respect to the number of lottery tickets. In addition, we provide an efficient algorithm for a more general setting in which the ticket price is constrained. To the best of the authors' knowledge, this is the first study that employs the CPT framework for designing an optimal lottery.