In econometrics, the truncated normal hurdle model is a variant of the Tobit model and was first proposed by Cragg in 1971. In a standard Tobit model, represented as y = ( x β + u ) 1 [ x β + u > 0 ] {\displaystyle y=(x\beta +u)1[x\beta +u>0]} , where u | x ∼ N ( 0 , σ 2 ) {\displaystyle u|x\sim N(0,\sigma ^{2})} This model construction implicitly imposes two first order assumptions:
Since: ∂ P [ y > 0 ] / ∂ x j = φ ( x β / σ ) β j / σ {\displaystyle \partial P[y>0]/\partial x_{j}=\varphi (x\beta /\sigma )\beta _{j}/\sigma } and ∂ E [ y ∣ x , y > 0 ] / ∂ x j = β j { 1 − θ ( x β / σ } {\displaystyle \partial \operatorname {E} [y\mid x,y>0]/\partial x_{j}=\beta _{j}\{1-\theta (x\beta /\sigma \}} , the partial effect of x j {\displaystyle x_{j}} on the probability P [ y > 0 ] {\displaystyle P[y>0]} and the conditional expectation: E [ y ∣ x , y > 0 ] {\displaystyle \operatorname {E} [y\mid x,y>0]} has the same sign: The relative effects of x h {\displaystyle x_{h}} and x j {\displaystyle x_{j}} on P [ y > 0 ] {\displaystyle P[y>0]} and E [ y ∣ x , y > 0 ] {\displaystyle \operatorname {E} [y\mid x,y>0]} are identical, i.e.:
∂ P [ y > 0 ] / ∂ x h ∂ P [ y > 0 ] / ∂ x j = ∂ E [ y ∣ x , y > 0 ] / ∂ x h ∂ E [ y ∣ x , y > 0 ] / ∂ x j = β h β j | {\displaystyle {\frac {\partial P[y>0]/\partial x_{h}}{\partial P[y>0]/\partial x_{j}}}={\frac {\partial \operatorname {E} [y\mid x,y>0]/\partial x_{h}}{\partial \operatorname {E} [y\mid x,y>0]/\partial x_{j}}}={\frac {\beta _{h}}{\beta _{j}}}|}
However, these two implicit assumptions are too strong and inconsistent with many contexts in economics. For instance, when we need to decide whether to invest and build a factory, the construction cost might be more influential than the product price; but once we have already built the factory, the product price is definitely more influential to the revenue. Hence, the implicit assumption (2) doesn't match this context. The essence of this issue is that the standard Tobit implicitly models a very strong link between the participation decision ( y = 0 {\displaystyle (y=0} or y > 0 ) {\displaystyle y>0)} and the amount decision (the magnitude of y {\displaystyle y} when y > 0 {\displaystyle y>0} ). If a corner solution model is represented in a general form: y = s ⋅ w , {\displaystyle y=s\centerdot w,} , where s {\displaystyle s} is the participate decision and w {\displaystyle w} is the amount decision, standard Tobit model assumes:
… excerpt ends here. Continue reading the full article.
