The Empirical Content of Binary Choice Models

An important goal of empirical demand analysis is choice and welfare prediction on counterfactual budget sets arising from potential policy interventions. Such predictions are more credible when made without arbitrary functional?form/distributional assumptions, instead based solely economic rationality, that is, consistent with utility maximization by a heterogeneous population. This paper investigates nonparametric rationality in the empirically context binary choice. We show under general unobserved heterogeneity, equivalent to pair Slutsky?like shape restrictions choice?probability functions. The forms these differ Slutsky inequalities for continuous goods. Unlike McFadden–Richter's stochastic revealed preference, our (a) global, their do not depend which how many observed, (b) closed form, hence easy impose parametric/semi/nonparametric models practical applications, (c) provide computationally simple, theory?consistent bounds budge sets.