Econometrics

econ.EM

Econometric theory, micro and macro econometrics, and empirical methods.

Sort:
Economics

Audit Cuts Green Patent Count by a Quarter

An error-as-signal audit combines PatentSBERTa disagreements with two-model climate-function judgments, reducing the Y02 estimate from 592,387 to 441,468 patents.

econ.EM
Aug 27, 20265 min2608.23420
Economics

Debiased Ridge Estimates Causal Effects in Panel Data

Individual ridge regressions approximate nonseparable potential outcomes while reducing shrinkage bias in large-T panels with many price controls.

econ.EM
Aug 1, 20265 min2607.28291
Economics

Latent Propensity Scores Expose Entry Bias in Demand Estimation

A finite-mixture correction models hidden market types across 19,544 airline markets, yielding higher price elasticities and lower inferred markups than 2SLS.

econ.EM
Jul 31, 20265 min2308.14196
Economics

Inventory A/B Tests Show Predictable Interference Bias

A pairwise randomization over items and time separates carryover from capacity crowding, reducing bias mechanisms that affect switchback and item-level designs.

econ.EM
Jul 29, 20265 min2501.11996
Economics

Polyads Reduce Bias in Sparse Multi-Way Networks

Conditioning on local polyad orbits removes fixed effects, delivering near-95% coverage where PPML intervals collapse in sparse simulations.

econ.EM
Jul 29, 20263 min2512.02203 Code available
Economics

Bias-Corrected LLM Choices Improve Economic Inference

AEM learns a correction from small human samples, matching a regional experiment’s -60 bps effect with -65±10 bps on hold-out regions.

econ.EM
Jul 28, 20265 min2510.25743
Economics

Outlier Choices Rarely Move Means but Flip Marginal Decisions

A pre-registered reanalysis of 358 behavioral-science meta-analyses shows median mean shifts of at most 0.047 Cohen’s d but decision reversals up to 15.9%.

econ.EM
Jul 28, 20263 min2607.23174
Economics

Triadic Closure Reframes Student Group Assignment Effects

A subgraph generated model separates dyadic links from triangles, showing social groups add 32 points to friendship and 3.7 points to triad formation.

econ.EM
Jul 28, 20265 min2607.17926 Code available
Economics

Local Threshold Tests Ignore Cross Dependence Asymptotically

Studentized local-linear jump estimates yield uniform existence and homogeneity tests, with simulations reaching 0.97 power when 20% of series jump.

econ.EM
Jul 28, 20264 min2312.01162
EconomicsPick

Auction Design for LLM Marketplaces Achieves Near-Optimal Welfare

A modified second-price mechanism ensures truthful quality reporting from competing language model providers.

econ.EMecon.TH
Jun 5, 20249 min2406.01234