I4R Working Paper: Instrument-Hacking Induces Median Bias in IV Estimators
RexDouglass · x · 2026-09-05
A new I4R discussion paper by Keane, Neal and Vu, Instrument-Hacking, shows that when researchers evaluate multiple candidate instruments and selectively report the specification with the best first- or second-stage statistics, IV estimators acquire median bias toward the OLS estimand, undermining the rationale for using IV.
- When all candidate instruments have the same true strength, median bias increases monotonically with the number of available instruments.
- When strengths differ, monotonicity need not hold, since more candidates can help researchers pick genuinely stronger instruments.
- However, simulations calibrated to the empirical distribution of instrument strengths in the IV literature show monotonic increases in bias, and the bias is substantial even with only a few instruments.
Topics covered include 2SLS, GMM, causal inference and size distortion.
More from Research
- Claude autonomously produces first computer-checked proof of Fermat's Last Theorem in 11 days — hornof · 2026-09-05
- Principia benchmark: top video models score ~0.8 on VBench but under 0.42 on physical consistency — anand_bhattad · 2026-09-05
- Terence Tao responds to rumors that an AI lab cracked the Navier-Stokes problem — elsleightholm · 2026-09-05
- New overview and evaluation of double robust flexible adjustment methods for causal inference — RexDouglass · 2026-09-05
- GPC Opensources Flow-Matching Robot Policies Trained via Sampling-Based Predictive Control — rsasaki0109 · 2026-09-05
- Blog Series Revisits AdaGrad, Reproducing Full Derivation via Upper Bound Minimization — aaron_defazio · 2026-09-05