A continuous treatment model for estimating a Dose Response Function under endogeneity and heterogeneous response to observable confounders: Description and implementation via the Stata module “ctreatreg”

This paper presents an original econometric model for estimating a Dose Response Function though a regression approach when treatment is continuous, individuals may react heterogeneously to observable confounders and selection-into-treatment may be (potentially) endogenous. After the description of the model, two estimation procedures are set out: one based on OLS under conditional mean independence, and one based on IV under selection endogeneity. The paper goes on by presenting ctreatreg, a author’s user-written Stata routine for an easy implementation of such a model. The paper proceeds by performing a Monte Carlo experiment to test the reliability of the model and of its associated Stata routine. Results show that the model and the Stata routine ctreatreg are both reliable as estimates consistently fit the expected results.