Propensity Score Modeling: Key Challenges When Moving Beyond the No-Interference Assumption

TL;DR

This paper explores the challenges of propensity score modeling in causal inference when the no-interference assumption is violated, proposing models based on mixed effects to address these issues.

Contribution

It identifies key challenges and introduces new propensity score models tailored for settings with interference, expanding causal inference methods.

Findings

Highlighting challenges in non-no-interference settings

Proposing mixed effects-based propensity score models

Laying groundwork for future research in interference scenarios

Abstract

The paper presents some models for the propensity score. Considerable attention is given to a recently popular, but relatively under-explored setting in causal inference where the no-interference assumption does not hold. We lay out some key challenges in propensity score modeling under interference and present a few promising models based on existing works on mixed effects models.