Estimation of Censored Demand Equations from Large Cross-Section Data

The article develops an alternative econometric methodology to estimate a system of censored demand equations using a large cross-section data from Colombian urban households. The approach preserves the behavioral information expressed by zero expenditures and conforms with the requirements imposed by consumer theory in a way consistent with the random utility hypothesis. We motivate the choice of the Tobit model as a statistical representation of consumer behavior and introduce the methodology by specifying the AIDS model modified according to both a translating and scaling demographic transformation. We propose to estimate each demand equation in unrestricted form using the jackknife technique. We then recover the demand parameters imposing the cross-equations restrictions by using minimum distance estimation. The empirical results of the censored demand system for specific households of policy relevance are reported