arXiv 14 Oct 2023 · Statistics — Methodology · publishedBiometrika (2025) · 1 citations (OpenAlex)
arXiv:2310.09545 · PDF · DOI · OpenAlex · Extracted main text
Recently, there has been a surge in methodological development for the difference-in-differences (DiD) approach to evaluate causal effects. Standard methods in the literature rely on the parallel trends assumption to identify the average treatment effect on the treated. However, the parallel trends assumption may be violated in the presence of unmeasured confounding, and the average treatment effect on the treated may not be useful in learning a treatment assignment policy for the entire population. In this article, we propose a general instrumented DiD approach for learning the optimal treatment policy. Specifically, we establish identification results using a binary instrumental variable (IV) when the parallel trends assumption fails to hold. Additionally, we construct a Wald estimator, novel inverse probability weighting (IPW) estimators, and a class of semiparametric efficient and multiply robust estimators, with theoretical guarantees on consistency and asymptotic normality, even when relying on flexible machine learning algorithms for nuisance parameters estimation. Furthermore, we extend the instrumented DiD to the panel data setting. We evaluate our methods in extensive simulations and a real data application.
appendix boundary found by appendix_command · 39% of the source is main text. Read the extracted text to check this.
The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.
| Reference | Intensity | Mentions | Sections | Main text | |
|---|---|---|---|---|---|
| 1 | Yifan Cui and Eric Tchetgen Tchetgen (2021) A semiparametric instrumental variable approach to optimal treatment regimes under endogeneity self | 0.928 | 4 | 3 | 100% |
| 2 | Ting Ye, Ashkan Ertefaie, James Flory, Sean Hennessy, and Dylan S Sm… (2022) Instrumented difference-in-differences | 0.894 | 7 | 6 | 71% |
| 3 | Alberto Abadie (2005) Semiparametric difference-in-differences estimators | 0.843 | 3 | 3 | 100% |
| 4 | Susan Athey and Stefan Wager (2021) Policy learning with observational data | 0.737 | 3 | 2 | 100% |
| 5 | Liangjun Su, Irina Murtazashvili, and Aman Ullah (2013) Local linear gmm estimation of functional coefficient iv models with an application to estimating the rate of return to schooling | 0.644 | 4 | 2 | 50% |
| 6 | Yichun Hu, Nathan Kallus, and Xiaojie Mao (2022) Fast rates for contextual linear optimization | 0.644 | 2 | 2 | 100% |
| 7 | Alexander R Luedtke and Mark J van der Laan (2016) Statistical inference for the mean outcome under a possibly non-unique optimal treatment strategy | 0.644 | 2 | 2 | 100% |
| 8 | Alexander Luedtke and Antoine Chambaz (2020) Performance guarantees for policy learning | 0.644 | 2 | 2 | 100% |
| 9 | Zhengling Qi, Rui Miao, and Xiaoke Zhang (2023) Proximal learning for individualized treatment regimes under unmeasured confounding | 0.644 | 2 | 2 | 100% |
| 10 | Tat-Thang Vo, Ting Ye, Ashkan Ertefaie, Samrat Roy, James Flory, Sea… (2022) Structural mean models for instrumented difference-in-differences | 0.644 | 2 | 2 | 100% |
Showing the top 10 of 76 scored citations.
arXiv econ.EM papers that cite this one, ranked by how heavily they lean on it.
| Citing paper | Intensity | Mentions | Sections | |
|---|---|---|---|---|
| 1 | On a Debiased and Semiparametric Efficient Changes-in-Changes Estimator | 0.405 | 1 | 1 |
| 2 | Policy Learning with $$-Expected Welfare | 0.000 | 1 | 1 |