Food Demand in Mexico : An Application of the Amemiya-Tobin Approach to the Estimation of a Censored Food System

The modeling of micro-level food demand patterns requires not only allowing for household heterogeneity, but also addressing the problem of censoring. In this article, we present a variation of the Amemiya-Tobin framework for estimating a censored demand system that allows for household heterogeneity. The unique aspect of our approach is the use of a procedure that ensures the adding up of both latent and observed expenditure shares and also imposes expenditure share nonnegativity. This system is applied to an analysis of food demand based on a random sample of urban Mexican households