2024-06-30-kalavasis24a.md

title	section	abstract	layout	series	publisher	issn	id	month	tex_title	firstpage	lastpage	page	order	cycles	bibtex_author	author	date	address	container-title	volume	genre	issued	pdf	extras
Smaller Confidence Intervals From IPW Estimators via Data-Dependent Coarsening (Extended Abstract)	Original Papers	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 $\epsilon>0$ 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 $\epsilon$-inaccurate propensity scores and $n$ samples, its confidence interval size scales with $\epsilon+(1/\sqrt{n})$. In contrast, under the same assumptions, existing estimators’ confidence interval sizes are $\Omega(1)$ irrespective of $\epsilon$ and $n$. Crucially, our estimator is data-dependent and we show that no data-independent CIPW estimator can be robust to inaccuracies.	inproceedings	Proceedings of Machine Learning Research	PMLR	2640-3498	kalavasis24a	0	Smaller Confidence Intervals From IPW Estimators via Data-Dependent Coarsening (Extended Abstract)	2767	2767	2767-2767	2767	false	Kalavasis, Alkis and Mehrotra, Anay and Zampetakis, Manolis	given family Alkis Kalavasis given family Anay Mehrotra given family Manolis Zampetakis	2024-06-30		Proceedings of Thirty Seventh Conference on Learning Theory	247	inproceedings	date-parts 2024 6 30	https://proceedings.mlr.press/v247/kalavasis24a/kalavasis24a.pdf