Benchmarking hidden confounders against covariates and pre-trends to weigh explanations

analisereal · x · 2026-09-17

The framework lets researchers benchmark unobserved confounders in two ways: as strong as observed covariates (red), or as strong as needed to generate the observed pre-trends (blue). This helps assess the plausibility of such confounders and adjudicate between competing explanations.

Related event: New framework quantifies confounding strength in DiD sensitivity analysis(4 posts)→

Original post →

More from Research

Research channel →