Phillip Heiler from Aarthus will present:
"Estimating Heterogeneous Bounds for Treatment Effects under Sample Selection and Non-response"
In this paper we propose a method for nonparametric identification and estimation
of heterogeneous bounds for causal effect parameters in general sample selection
models where the initial treatment can affect whether an post-intervention outcome
is observed or not. The original treatment selection can be confounded by observable
covariates while the outcome selection can be affected by both observables
and unobservables. The method provides functional estimates of conditional effect
bounds dependent on pre-treatment characteristics. It allows for
conducting valid statistical inference on the unidentified conditional effect curves.
We use a flexible semiparametric de-biased machine learning approach that can
accommodate flexible functional forms and high-dimensional set of observed confounding
variables in both treatment and outcome selection process.