Alkis Kalavasis, Anay Mehrotra, Manolis Zampetakis
arXiv 2 Oct 2024 · Statistics — Methodology
arXiv:2410.01658 · PDF · DOI · OpenAlex · Extracted main text
Inverse propensity-score weighted (IPW) estimators are prevalent in causal inference for estimating average treatment effects in observational studies. Under unconfoundedness, given accurate propensity scores and $n$ samples, the size of confidence intervals of IPW estimators scales down with $n$, and, several of their variants improve the rate of scaling. However, neither IPW estimators nor their variants are robust to inaccuracies: even if a single covariate has an $\varepsilon>0$ additive error in the propensity score, the size of confidence intervals of these estimators can increase arbitrarily. Moreover, even without errors, the rate with which the confidence intervals of these estimators go to zero with $n$ can be arbitrarily slow in the presence of extreme propensity scores (those close to 0 or 1). We introduce a family of Coarse IPW (CIPW) estimators that captures existing IPW estimators and their variants. Each CIPW estimator is an IPW estimator on a coarsened covariate space, where certain covariates are merged. Under mild assumptions, e.g., Lipschitzness in expected outcomes and sparsity of extreme propensity scores, we give an efficient algorithm to find a robust estimator: given $\varepsilon$-inaccurate propensity scores and $n$ samples, its confidence interval size scales with $\varepsilon+1/\sqrt{n}$. In contrast, under the same assumptions, existing estimators' confidence interval sizes are $\Omega(1)$ irrespective of $\varepsilon$ and $n$. Crucially, our estimator is data-dependent and we show that no data-independent CIPW estimator can be robust to inaccuracies.
appendix boundary found by appendix_command · 91% 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 | Hernan, Miguel A., Robins, James M (2023) Causal Inference: What If | 0.928 | 4 | 3 | 100% |
| 2 | Imbens, Guido W., Rubin, Donald B (2015) Causal Inference for Statistics, Social, and Biomedical Sciences: An Introduction | 0.894 | 7 | 4 | 71% |
| 3 | Wager, Stefan, Athey, Susan (2018) Estimation and Inference of Heterogeneous Treatment Effects using Random Forests | 0.811 | 4 | 2 | 100% |
| 4 | McCaffrey, Daniel F., Ridgeway, Greg, Morral, Andrew R (2004) Propensity Score Estimation With Boosted Regression for Evaluating Causal Effects in Observational Studies | 0.737 | 3 | 2 | 100% |
| 5 | Wager, Stefan (2020) STATS 361: Causal Inference | 0.737 | 3 | 2 | 100% |
| 6 | Foster, Dylan J., Syrgkanis, Vasilis (2023) Orthogonal Statistical Learning | 0.644 | 4 | 2 | 50% |
| 7 | Hirano, Keisuke, Imbens, Guido W., Ridder, Geert (2003) Efficient Estimation of Average Treatment Effects Using the Estimated Propensity Score | 0.644 | 2 | 2 | 100% |
| 8 | Shalev-Shwartz, Shai, Ben-David, Shai (2014) Understanding Machine Learning: From Theory to Algorithms | 0.644 | 2 | 2 | 100% |
| 9 | Chernozhukov, Victor, Chetverikov, Denis, Demirer, Mert, Duflo, Esth… (2018) Double/Debiased Machine Learning for Treatment and Structural Parameters | 0.511 | 3 | 2 | 33% |
| 10 | Bang, Heejung, Robins, James M (2005) Doubly Robust Estimation in Missing Data and Causal Inference Models | 0.511 | 3 | 2 | 33% |
Showing the top 10 of 48 scored citations.