Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
The purpose of this paper is to emphasize the importance of double-hurdle models in the estimation of tobacco demand equations. Using data from the Spanish Family Expenditure Survey (EPF), Tobit, P-Tobit, and first hurdle dominance models are shown to be restrictive. This dataset also confirms the existence of separate individual decisions on participation and consumption. The authors propose to estimate a dependent version of a double-hurdle model, although independence is not restrictive once additional powers of total expenditure are included in the specification of the second hurdle. Several misspecification tests are also conducted among the different models. Copyright 1996 by Blackwell Publishing Ltd