My paper Lottery-spending: a non-parametric analysis has appeared. This was written jointly with undergrads Kayla Frisoli, Li Ke, and Melody Lim, and describes the results of a project we did last spring. The problem is to understand those people who spend a lot on lottery tickets and how much they actually spend. We made use of publicly-available surveys of lottery participation and used non-parametric methods to study the distribution. (Previous studies had used regression.)
We studied the heavy tail of the distribution and the variation in the distribution from survey to survey, and we compared the distribution for various demographic groups. We found that the observed higher spending by Hispanic lottery players can be attributed to differences in education levels, and our results contradict previous claims that the top 10% of lottery players consistently account for 50% of lottery sales.