Masahiro Kato, Masaaki Imaizumi
arXiv 10 Feb 2022 · Econometrics · 3 citations (OpenAlex)
arXiv:2202.05245 · PDF · DOI · OpenAlex · Extracted main text
We study the benign overfitting theory in the prediction of the conditional average treatment effect (CATE), with linear regression models. As the development of machine learning for causal inference, a wide range of large-scale models for causality are gaining attention. One problem is that suspicions have been raised that the large-scale models are prone to overfitting to observations with sample selection, hence the large models may not be suitable for causal prediction. In this study, to resolve the suspicious, we investigate on the validity of causal inference methods for overparameterized models, by applying the recent theory of benign overfitting (Bartlett et al., 2020). Specifically, we consider samples whose distribution switches depending on an assignment rule, and study the prediction of CATE with linear models whose dimension diverges to infinity. We focus on two methods: the T-learner, which based on a difference between separately constructed estimators with each treatment group, and the inverse probability weight (IPW)-learner, which solves another regression problem approximated by a propensity score. In both methods, the estimator consists of interpolators that fit the samples perfectly. As a result, we show that the T-learner fails to achieve the consistency except the random assignment, while the IPW-learner converges the risk to zero if the propensity score is known. This difference stems from that the T-learner is unable to preserve eigenspaces of the covariances, which is necessary for benign overfitting in the overparameterized setting. Our result provides new insights into the usage of causal inference methods in the overparameterizated setting, in particular, doubly robust estimators.
appendix boundary found by appendix_command · 38% 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 | Bartlett, P. L., Long, P. M., Lugosi, G., and Tsigler, A (2020) Benign overfitting in linear regression | 0.754 | 63 | 11 | 43% |
| 2 | Kennedy, E. H (2020) Optimal doubly robust estimation of heterogeneous causal effects | 0.737 | 3 | 2 | 100% |
| 3 | Künzel, S. R., Sekhon, J. S., Bickel, P. J., and Yu, B (2019) Metalearners for estimating heterogeneous treatment effects using machine learning | 0.644 | 4 | 1 | 100% |
| 4 | Abrevaya, J., Hsu, Y.-C., and Lieli, R. P (2015) Estimating Conditional Average Treatment Effects | 0.644 | 2 | 2 | 100% |
| 5 | Chernozhukov, V., Chetverikov, D., Demirer, M., Duflo, E., Hansen, C… (2018) Double/debiased machine learning for treatment and structural parameters | 0.644 | 2 | 2 | 100% |
| 6 | Hahn, J (1998) On the Role of the Propensity Score in Efficient Semiparametric Estimation of Average Treatment Effects | 0.644 | 2 | 2 | 100% |
| 7 | Imbens, G. W. and Rubin, D. B (2015) Causal Inference for Statistics, Social, and Biomedical Sciences: An Introduction | 0.644 | 2 | 2 | 100% |
| 8 | Koltchinskii, V. and Lounici, K (2017) Concentration inequalities and moment bounds for sample covariance operators | 0.585 | 3 | 3 | 33% |
| 9 | Nie, X. and Wager, S (2020) Quasi-Oracle Estimation of Heterogeneous Treatment Effects | 0.585 | 3 | 1 | 100% |
| 10 | Assmann, S., Pocock, S., Enos, L., and Kasten, L (2000) Subgroup analysis and other (mis)uses of baseline data in clinical trials | 0.511 | 2 | 1 | 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 | Double and Single Descent in Causal Inference with an Application to High-Dimensional Synthetic Control | 0.644 | 2 | 2 |