A Multivariate Sample‐Selection Model: Estimating Cigarette and Alcohol Demands with Zero Observations

摘要: Heckman's sample-selection model is generalized to the multivariate case and used estimate demands for cigarettes, beer, wine by individuals in United States. The specification allows correlations between error terms of multiple selection level equations, nests popular bivariate two-part model. Empirical results suggest that proposed performs better than restricted specifications. Differentiated effects variables on probabilities levels consumption also rejection Tobit system. Gender differences are present, demographic more important income determining wine.