A Meta-learner for Heterogeneous Effects in Difference-in-Differences
Hui Lan, Haoge Chang, Eleanor Dillon, Vasilis Syrgkanis · 2025
Read A Meta-learner for Heterogeneous Effects in Difference-in-Differences by Hui Lan, Haoge Chang, Eleanor Dillon, Vasilis Syrgkanis (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2502-04699.txt.
Description
Hui Lan, Haoge Chang, Eleanor Dillon, and Vasilis Syrgkanis build an estimator for varied treatment effects in panel data in a 2025 paper. The difference-in-differences design compares treated and control groups before and after a policy, assuming parallel paths without it. Effects may differ across subgroups, and estimating that variation flexibly is hard. Their meta-learner targets the average effect on the treated for each profile.
The method reduces the problem to a single convex fitting task over helper models. A robustness property shields it from errors in those helpers. Extensions cover covariate shift and non-compliance settings. Tests show it beats existing baselines, the authors report.
Catalog record
| CREATOR | Hui Lan, Haoge Chang, Eleanor Dillon, Vasilis Syrgkanis |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 106 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2502-04699