Abstract:This paper
presents ctreatreg, a Stata module for estimating a dose-response
function when: (i) treatment is continuous, (ii) individuals may react
heterogeneously to observable confounders, and (iii) selection-into-treatment
may be endogenous. Two estimation procedures are implemented: OLS under
Conditional Mean Independence, and Instrumental-Variables (IV) under selection
endogeneity. A Monte Carlo experiment to test the reliability of the proposed
command is finally set out.
Keywords:Stata commands; treatment effects, dose-response function, continuous
treatment, Monte Carlo, R&D support