We investigate the impact of income distribution on lottery expenditures in Taiwan, using an agent-based model developed in [2, 3]. The agents in the model are potential lottery buyers, whose characteristics are described by three features: the percentage of income spent on the lottery, the preferences among lottery numbers selected and the aversion to regret. We used a genetic algorithm to drive the model simulation under agents with different incomes, based on household income data in Taiwan from 1979 to 2003. The simulation results indicated that the impact of income distribution on lottery sales is not significant. This might be due to the Taiwan economy having a minor degree of income variation which has a low effect on lottery expenditures.
Natural Computing in Computational Finance Studies in Computational Intelligence Volume 185, 2009, pp 207-223