Estimating Heterogeneous Treatment Effects with Confounded Instruments
Editor assessment Waiting for Editors 6d left- Editor
- Unassigned
- Submitted
- Nov 4, 2026
- Due
- Nov 11, 2026
- Reviews
- 0 of 0
- Status
- Waiting for Editors
Assign an editor to this article
Choose a handling editor to move this submission into the active editorial queue.
Abstract
We show that instrumental variables estimates of heterogeneous treatment effects are biased when the instrument is correlated with effect modifiers. We propose a two-stage semiparametric correction and establish root-n consistency under mild regularity conditions.