📋 Tender Overview
High-Dimensional Nonparametric Bayesian Causal Inference
Bayesian Nonparametric Approach For Variable Selection In High-Dimensional Causal Inferencecausal Inference Can Become Harder When More Data Are Available. In High-Dimensional Settings, The Inclusion Of All Variables Is Impossible, While The Inclusion Of Too Few Leads To Results That Can Be Wrong. Thus, Variable Selection Is A Necessity, Yet Available Methods Are Limited. The Erc-Funded Baycause Project Aims To Develop Bayesian Nonparametric Methods And Theories Suitable For Variable Selection In High-Dimensional Causal Inference. Recent Theoretical Advances In Bernstein-Von Mises Theory And High-Dimensional Nonparametric Regression Enabled The Application Of Bayesian Nonparametric Approaches In Causal Inference. The Methods Developed By The Project Will Expand The Limited Settings In Which Trustworthy High-Dimensional Causal Inference Is Possible, Leading To Applications In Medicine, Economics And Other Fields.
Funded Under
European Research Council (Erc)
Total Cost
€ 1 596 136,25
Eu Contribution
€ 1 499 770,00
Coordinated By
Stichting Vumc
Netherlands
