On Asymptotically Efficient Estimation in Semiparametric Models

By Anton Schick *. 1986

Subfields: Statistics and Probability, Statistical Learning Theory

How researchers see this paper

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

In Statistics and Probability

In Statistical Learning Theory

Citations in the corpus

Double/De-Biased Machine Learning of Global and Local Parameters Using Regularized Riesz Representers (2018, background): “See, e.g, Schick (1986) for early use and Chernozhukov et al. (2018a) for more recent use in the context of debiased machine learning.”
Generalized random forests (2016, neutral): “2014/10/16 file: paper.tex date: April 6, 2018 (Chernozhukov et al., 2016; Schick, 1986).”
Learning When-to-Treat Policies (2019, positive): “8The idea of cross-fitting has gained growing popularity recently to reduce the effect of own-observation bias and to enable results on semiparametric rates of convergence using generic nuisance component estimates (Schick 1986; Chernozhukov et al. 2018; Athey and Wager 2020).”
Efficient Policy Learning (2017, methodology): “The K-fold algorithmic structure used in (9) was proposed by Schick (1986) as a general purpose tool for efficient estimation in semiparametric models, and was independently called “cross-fitting” by Chernozhukov et al. (2016) and “cross-estimation” by Wager et al. (2016).”
Quasi-oracle estimation of heterogeneous treatment effects (2017, background): “Then, and so (cid:107) T as well as feasible analogues obtained by cross-ﬁtting (Chernozhukov et al., 2017; Schick, 1986): where q is a mapping from the i = 1 , ..., n sample indices to Q evenly sized data folds, such that ˆ e ( − q ( i )) ( x ) and ˆ m ( − q ( i )) ( x ) are each trained without…”
Policy Learning With Observational Data (2017, methodology): “…ﬁrst estimating g ( · ) and m ( · ), and They show that this estimator is √ n -consistent and asymptotically unbiased Gaussian for θ , provided that the nuisance estimates ˆ g ( · ) and ˆ m ( · ) converge suﬃciently fast and that we use cross-ﬁtting (Chernozhukov et al., 2018a; Schick, 1986).”
On Adaptive Estimation in Nonstationary Arma Models with Garch Errors (2003, methodology): “By parameterizing the density, Koul and Schick (1996, 1997) gave some clear explanations as to the adaptivity of the parameters in the random AR and nonlinear AR models. This technique requiring the parameterization of densities is discussed carefully in BKRW (1993). As in Koul and Schick (1996, 199”
Offline Multi-Action Policy Learning: Generalization and Optimization (2018, methodology): “The second component of our approach, using the K-fold structure, builds on a commonly employed method in statistical inference (see Schick (1986), Chernozhukov et al. (2016a), Athey and Wager (2017)) to reduce model overfit.”
Semiparametric doubly robust targeted double machine learning: a review (2022, background): “The latter advantage seems to have driven its initial use in functional estimation problems, as in, e.g., Hasminskii and Ibragimov [1978], Pfanzagl [1982], Schick [1986], Bickel and Ritov [1988], etc.”
A unifying approach for doubly-robust $\ell_1$ regularized estimation of causal contrasts (2019, background): “1, but with the stringent requirement that the approximation error r (Z) be equal to 0 has been assumed by Tan (2018) and Ning et al.”
Sharp Sensitivity Analysis for Inverse Propensity Weighting via Quantile Balancing (2021, unknown): “The specific form of sample splitting analyzed in our proofs is “cross-fitting” (Schick 1986; Newey and Robins 2017; Chernozhukov et al. 2018), but leave-one-out or out-of-bag quantile estimates perform similarly in simulations.”