Heterogeneity and the nonparametric analysis of consumer choice: conditions for invertibility”, cemmap Working Papers

This paper considers structural nonparametric random utility models for continuous choice variables with unobserved heterogeneity. We provide sufficient conditions on ran-dom preferences to yield reduced-form systems of nonparametric stochastic demand func-tions that allow global invertibility between demands and nonseparable unobserved hetero-geneity. We distinguish between new classes of models in which heterogeneity is separable and nonseparable in the marginal rates of substitution, respectively. Invertibility is essen-tial for global identification of structural consumer demand models, for the existence of well-specified probability models of choice and for the nonparametric analysis of revealed stochastic preference