Treatment effect in a proportional hazards model with instrumental variables
Principal Investigator
Name
lorenzo tedesco
Degrees
M.D.
Institution
KU LEUVEN
Position Title
PhD student
Email
About this CDAS Project
Study
HIPB
(Learn more about this study)
Project ID
HIPB-10
Initial CDAS Request Approval
Dec 5, 2022
Title
Treatment effect in a proportional hazards model with instrumental variables
Summary
In this paper we consider the Cox’s proportional hazards model, proposing a novel approach for the identification and estimation of the causal hazard ratio in the presence of unmeasured confounding factors and random right censoring.
The approach is based on a discrete instrumental variable and allows to include a discrete endogenous covariate and any type of exogenous covariate. A non-parametric instrumental regression for right censored duration outcomes is used to estimate the parameter of interest of the underlying semiparametric proportional hazards model. Asymptotic properties of the estimator are derived. The approach is illustrated via simulation studies. We would like to use your data for an emepirical application.
The approach is based on a discrete instrumental variable and allows to include a discrete endogenous covariate and any type of exogenous covariate. A non-parametric instrumental regression for right censored duration outcomes is used to estimate the parameter of interest of the underlying semiparametric proportional hazards model. Asymptotic properties of the estimator are derived. The approach is illustrated via simulation studies. We would like to use your data for an emepirical application.
Aims
- Usage of the data for example application of a new statistical method
Collaborators
Professor Ingrid Van Keilegom