Learning Objectives for Treatment Effect Estimation

By Xinkun Nie, Stefan Wager. 2017

67 citations recorded by the source corpus.

How researchers see this paper

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

Citations in the corpus

Local Linear Forests (2018, methodology): “From the perspective, we can see local linear causal forests as an adapted R-learner method; note that Nie and Wager (2017) and Kennedy (2020) give quasi-oracle and double-robustness properties of R-learner estimates.”
Metalearners for estimating heterogeneous treatment effects using machine learning (2017, background): “[33] ≤ L(2)c̃ ∗ ( 1 b n k1 c 2/d + 1 b2 k2 c 2/d ) ≤ 2c̃L(2) ∗ (( k1 n )2/d + ( k0 m )2/d) .”
Efficient Policy Learning (2017, background): “Finally, we also note a growing literature on estimating conditional average treatment effects (Athey and Imbens, 2016; Athey, Tibshirani, and Wager, 2018; Chen and Pouzo, 2012; Nie and Wager, 2017; Künzel, Sekhon, Bickel, and Yu, 2017).”
Assessing Treatment Effect Variation in Observational Studies: Results from a Data Challenge (2019, methodology): “Although the original estimator (4) was designed for constant treatment effect estimation, Nie and Wager (2017) showed that we can use it to motivate an “R-learner” objective function for heterogeneous treatment effect estimation, τ̂(·) = argminτ { n ∑ i=1 (( Yi − m̂(Xi) ) − τ(Xi) ( Zi − ê(Xi) ))2 +”
Augmented minimax linear estimation (2017, methodology): “We then fit the τ(x) function via the R-lasso method proposed by Nie and Wager (2017), again on φ(x), and finally set µ̂(x) = Ê [ Yi ∣∣Xi = x]− τ̂(x)ê(x).”
Estimating Treatment Effects with Causal Forests: An Application (2019, unknown): “One promising avenue to heterogeneous treatment eﬀect estimation starts from an early result of Robinson (1988) on inference in the partially linear model (Nie and Wager, 2017; Zhao et al., 2017).”
Confounding-Robust Policy Improvement (2018, methodology): “These methods, as well as approaches to predict individual-level causal effects from observational data (Künzel et al. 2017, Nie and Wager 2017, Shalit et al. 2017, Wager and Athey 2017b), operate under the controversial assumption of unconfoundedness, which requires that the data are sufficiently…”