no code implementations • 14 Apr 2024 • Jin-Hong Du, Zhenghao Zeng, Edward H. Kennedy, Larry Wasserman, Kathryn Roeder
In this paper, we propose a generic semiparametric inference framework for doubly robust estimation with multiple derived outcomes, which also encompasses the usual setting of multiple outcomes when the response of each unit is available.
1 code implementation • 22 Mar 2024 • Alec McClean, Sivaraman Balakrishnan, Edward H. Kennedy, Larry Wasserman
Then, assuming the nuisance functions are H\"{o}lder smooth, but without assuming knowledge of the true smoothness level or the covariate density, we establish that DCDR estimators with several linear smoothers are semiparametric efficient under minimal conditions and achieve fast convergence rates in the non-$\sqrt{n}$ regime.
no code implementations • 31 Jan 2024 • Zhenghao Zeng, David Arbour, Avi Feller, Raghavendra Addanki, Ryan Rossi, Ritwik Sinha, Edward H. Kennedy
In this paper, we study the role of surrogates in estimating continuous treatment effects and propose a doubly robust method to efficiently incorporate surrogates in the analysis, which uses both labeled and unlabeled data and does not suffer from the above selection bias problem.
no code implementations • 6 May 2023 • Sivaraman Balakrishnan, Edward H. Kennedy, Larry Wasserman
These first-order methods are however provably suboptimal in a minimax sense for functional estimation when the nuisance functions live in Holder-type function spaces.
no code implementations • 15 Jan 2023 • Kwangho Kim, Edward H. Kennedy, José R. Zubizarreta
We study counterfactual classification as a new tool for decision-making under hypothetical (contrary to fact) scenarios.
no code implementations • 19 Jul 2022 • Amanda Coston, Edward H. Kennedy
We provide a new definition of collapsibility that makes this choice of aggregation method explicit, and we demonstrate that the odds ratio is collapsible under geometric aggregation.
2 code implementations • 11 Mar 2021 • Ian Waudby-Smith, David Arbour, Ritwik Sinha, Edward H. Kennedy, Aaditya Ramdas
This paper introduces time-uniform analogues of such asymptotic confidence intervals, adding to the literature on confidence sequences (CS) -- sequences of confidence intervals that are uniformly valid over time -- which provide valid inference at arbitrary stopping times and incur no penalties for "peeking" at the data, unlike classical confidence intervals which require the sample size to be fixed in advance.
no code implementations • 2 Mar 2021 • Liu Leqi, Edward H. Kennedy
In this work, we propose a new median optimal treatment regime that instead treats individuals whose conditional median is higher under treatment.
no code implementations • NeurIPS 2020 • Amanda Coston, Edward H. Kennedy, Alexandra Chouldechova
We propose a doubly-robust procedure for learning counterfactual prediction models in this setting.
1 code implementation • 30 Aug 2019 • Amanda Coston, Alan Mishler, Edward H. Kennedy, Alexandra Chouldechova
These tools thus reflect risk under the historical policy, rather than under the different decision options that the tool is intended to inform.
1 code implementation • 9 Jul 2019 • Kwangho Kim, Edward H. Kennedy, Ashley I. Naimi
Modern longitudinal studies collect feature data at many timepoints, often of the same order of sample size.
no code implementations • 8 Jun 2018 • Kwangho Kim, Jisu Kim, Edward H. Kennedy
In this paper we develop a framework for characterizing causal effects via distributional distances.
2 code implementations • 20 Nov 2017 • Ashley I. Naimi, Edward H. Kennedy
We use 10, 000 simulated samples and 50, 100, 200, 600, and 1200 observations to investigate the bias and mean squared error of singly robust (g Computation, inverse probability weighting) and doubly robust (augmented inverse probability weighting, targeted maximum likelihood estimation) estimators under four scenarios: correct and incorrect model specification; and parametric and nonparametric estimation.
Methodology