Threshold Effects in Cigarette Addiction: An Application of the Threshold Model in Dynamic Panels

We adopt the threshold model of myopic cigarette addiction to US state-level panel data. The threshold model is used to identify the structural effects of cigarette demand determinants across the income stratification. Furthermore, we apply a bootstrap approach to correct for the small-sample bias that arises in the dynamic panel threshold model with fixed effects. Our empirical results indicate that there exists the heterogeneity of smoking dynamics across consumers.