NBER's fereg beats reghdfe by up to 204x; author built it with Codex and Claude
daveholtz · x · 2026-11-19
A new NBER working paper (w36017) by Johannes F. Schmieder introduces fereg and feppml, Stata commands for high-dimensional fixed-effects linear and Poisson regressions. A compiled plugin picks between projection, Krylov, and multigrid solvers, runs multi-core, and verifies every result.
- Up to 6–204x faster than reghdfe and 11–103x faster than ppmlhdfe on simulations with up to 10M observations
- Beat R's fixest and Python's PyFixest in all 53 simulated configurations, with geometric-mean time ratios of 3.0–7.7
- Supports interacted fixed effects, group-specific slopes, weights, robust and multiway clustered SEs
- The author discloses using OpenAI's Codex and Anthropic's Claude for development, testing, and writing
The quoting tweet jokes "we need to pace the frontier immediately" — even econometrics tooling is now being accelerated by AI.
More from Research
- SE3 Labs Releases First Open YAM Real-Robot Teleop Dataset: 500 Bimanual Episodes — paigeinsf · 2026-11-19
- 1.5 years under review: researchers urge releasing AI papers immediately after submission — carlosdponx · 2026-11-19
- Berkeley's StreamEmbed indexes your screen history on-device, 3.7x faster at matched accuracy — berkeley_ai · 2026-11-19
- NAACL RAF opens 2026-2027 grant proposals for NLP initiatives, due May 6 — keviv9 · 2026-11-19
- Open-sourced Indic profanity detector covers 12 Indian languages, runs at 4ms per comment on CPU — sudalairajkumar · 2026-11-19
- MAVE reconstructs editable 3D blockouts from film shots, works inside Blender — bdsqlsz · 2026-11-19