Revealed preference restrictions are increasingly used to bound demand responses and as shape restrictions in nonparametric estimation exercises. However, the restrictions imposed are not sufficient for rationality when predictions are made at more than a single price regime. We highlight the nonlinearities in revealed preference restrictions and the nonconvexities in the set of predictions that arise when making multiple predictions. We develop a mixed integer programming characterisation of the problem that can be used to impose rationality on multiple predictions. The approach is applied to the UK Family Expenditure Survey to recover jointly rational nonparametric estimates of income expansion paths.
Revealed Preference inequality restrictions set identification demand mixed integer programming