Journal article

Direct and indirect effects of continuous treatments based on generalized propensity score weighting


  • Huber, Martin ORCID Department of Economics, University of Fribourg, Fribourg, Switzerland
  • Hsu, Yu‐Chin Academia Sinica Institute of Economics, National Central University, Taoyuan City, Taiwan
  • Lee, Ying‐Ying Department of Economics, University of California Irvine, Irvine California, USA
  • Lettry, Layal Swiss Federal Agency for Social Insurances, Bern, Switzerland
Show more…
  • 2020
Published in:
  • Journal of Applied Econometrics. - Wiley. - 2020, vol. 35, no. 7, p. 814-840
English This paper proposes semi- and nonparametric methods for disentangling the total causal effect of a continuous treatment on an outcome variable into its natural direct effect and the indirect effect that operates through one or several intermediate variables called mediators jointly. Our approach is based on weighting observations by the inverse of two versions of the generalized
propensity score (GPS), namely the conditional density of treatment either given observed covariates or given covariates and the mediator. Our effect estimators are shown to be asymptotically normal when the GPS is estimated by either a parametric or a nonparametric kernel-based method. We also provide a simulation study and an empirical illustration based on the Job Corps experimental study.
Faculté des sciences économiques et sociales et du management
Département d'économie politique
  • English
Rights reserved
Open access status
Persistent URL

Document views: 0 File downloads:
  • jofappliedeconometrics-2020-huber-directandindirecteffectsofcontinuoustreatmentsbasedongeneralized.pdf: 0