New working paper quantifies omitted variable bias and robustness in DiD designs
analisereal · x · 2026-09-17
Carlos Cinelli, Juejue Wang, Pedro Sant'Anna and Victor Chernozhukov release a working paper, 'Omitted Variable Bias in Difference-in-Differences Designs.'
- Provides a framework quantifying OVB in DiD, yielding robustness values: how strong an unobserved confounder must be to overturn results.
- Benchmarks confounders against observed covariates or the strength needed to generate observed pre-trends, helping adjudicate competing explanations.
- Complements pre-trends benchmarking work (Rambachan & Roth) by explaining how parallel-trends violations could arise.
- All quantities are estimable nonparametrically with debiased ML via the dml.sensemakr R package; integration with standard DiD packages is coming.
Related event: Chernozhukov Team Releases Paper on Omitted Variable Bias in DID(2 posts)→
More from Research
- BuildingBench Tests Coding Agents at 3D Building Generation; Stealth Model Claims 10x Token Savings — Lianhuiq · 2026-09-17
- RoMem uses continuous phase rotation to expire stale agent memories, 2-3x better temporal reasoning — Roger_M_Taylor · 2026-09-17
- New Paper 'Instrumental Choices' Benchmarks When LLM Agents Take Policy-Violating Shortcuts — drbiomass · 2026-09-17
- APABOT: Open-source ParlAI chatbot prototype for autism spectrum conversation practice — AryHHAry · 2026-09-17
- LimiX-2 Debuts With Mechanism-Network Paradigm, Tops TabArena Tabular Benchmarks — stable-ai · 2026-09-17
- The Garden and the Tree: how AI will split mathematics into two distinct communities — yeastsplainer · 2026-09-17