Non‐parametric methods for doubly robust estimation of continuous treatment effects

By Edward H. Kennedy, Zongming Ma, Matthew McHugh, Dylan S. Small. 2015

328 citations recorded by the source corpus.

Continuous treatments (e.g. doses) arise often in practice, but many available causal effect estimators are limited by either requiring parametric models for the effect curve, or by not allowing doubly robust covariate adjustment. We develop a novel kernel smoothing approach that requires only mild smoothness assumptions on the effect curve and still allows for misspecification of either the treatment density or outcome regression. We derive asymptotic properties and give a procedure for data‐driven bandwidth selection. The methods are illustrated via simulation and in a study of the effect of nurse staffing on hospital readmissions penalties.

Subfields: Accounting, Finance

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Accounting

In Finance

Citations in the corpus

Double/De-Biased Machine Learning of Global and Local Parameters Using Regularized Riesz Representers (2018, methodology): “First, we use our method to answer a question in household finance: what is the average treatment effect of 401(k) eligibility on net financial assets (over a horizon of about two years)? We follow the identification strategy of Poterba and Venti (1994) and Poterba et al.”
Sequential Double Robustness in Right-Censored Longitudinal Models (2017, result): “We instead draw inspiration from the recent work of Kennedy et al. (2016), which gave an implementation of an infinite-dimensional targeting step in a continuous point treatment setting, where this step is implemented via locally linear regression.”
Double Debiased Machine Learning Nonparametric Inference with Continuous Treatments (2019, neutral): “preliminary element for policy learning and optimization with a continuous decision, following Manski (2004), Hirano and Porter (2009), Kitagawa and Tetenov (2018), Kallus and Zhou (2018), Demirer, Syrgkanis, Lewis, and Chernozhukov (2019), Athey and Wager (2019), Farrell, Liang, and Misra (2019), a”
Towards optimal doubly robust estimation of heterogeneous causal effects (2020, background): “This phenomenon was also noted in the dose-response estimation problem by Kennedy et al. [2017], and is in contrast with the error bounds given in Nie and Wager [2021] and Zimmert and Lechner [2019], which required ATE-like conditions for oracle efficiency of CATE estimators, via n − 1 / 4 or…”
Difference-in-Differences with a Continuous Treatment (2021, neutral): “[25] Edward H Kennedy, Zongming Ma, Matthew D McHugh, and Dylan S Small.”
Nonparametric estimation of causal heterogeneity under high-dimensional confounding (2019, positive): “For example efficient score based estimation can also be used for instrumental variables problems (Chernozhukov et al., 2018a), difference-in-differences estimation (Zimmert, 2018) and continuous treatment settings (Kennedy et al., 2017).”
The Balancing Act in Causal Inference (2021, methodology): “…of this approach can be used for other estimands like the dose-response curve for a continuous-valued treatment, µ(w0) = E[mw0(Xi)] = E[ ∫ mw(Xi)δw0(w)dw], which fix some aspects of the treatment or population and average over others (e.g., Chernozhukov, Newey, & Singh, 2018; Kennedy et al., 2017).”
Entropy Balancing for Continuous Treatments (2020, background): “2014; Yiu and Su 2018), machine learning (Colangelo and Lee 2020; Kennedy et al. 2017; Kreif et al. 2015; Su, Ura, and Zhang 2019; Zhu, Coffman, and Ghosh 2015) or a combination thereof.”
Semiparametric doubly robust targeted double machine learning: a review (2022, background): “These arise in many important and common settings in causal inference, for example: eﬀects of continuous treatments [Díaz and van der Laan, 2013, Kennedy et al., 2017, Rubin and van der Laan, 2006], counterfactual density estimation [Kennedy et al., 2021, Kim et al., 2018, Robins and Rotnitzky,…”