Lottery markets design, micro-structure, and macro-behavior: An ACE approach

An agent-based computational modeling of the lottery market is established in this paper to study the design issue, in terms of the lottery tax rate, as well as the emerging market behavior. By using genetic algorithms and fuzzy logic, lottery participants are modeled as autonomous agents who may endogenously adapt to exhibit behavioral properties consistent with well-noticed behavior of lottery markets. Three major findings are presented. First, as anticipated, a Laffer curve is found in this model; nonetheless, the Laffer curve has a flat top, which indicates the non-uniqueness of the optimal lottery tax rate. Second, conscious selection behavior is also observed, but it becomes weaker as time goes on. Third, for the halo effect, we observe exactly the opposite. Each of these three findings are then compared with available empirical results, and the mechanism of genetic algorithms is further examined in light of the anti-halo effect.