Nonparametric Analysis of Random Utility Models: Computational Tools for Statistical Testing

Kitamura and Stoye (2018) recently proposed a nonparametric statistical test for random utility models of consumer behavior. The is formulated in terms linear inequality constraints quadratic objective function. While the conceptually appealing, its practical implementation computationally challenging. In this paper, we develop column generation approach to operationalize test. These novel computational tools generate considerable gains practice, which substantially increases empirical usefulness Stoye's