EconBase
← All papers

Assumption-lean falsification tests of rate double-robustness of double-machine-learning estimators

Lin Liu, Rajarshi Mukherjee, James M. Robins

arXiv 18 Jun 2023 · Statistics — Methodology · publishedJournal of Econometrics (2023) · 6 citations (OpenAlex)

arXiv:2306.10590 · PDF · DOI · OpenAlex · Extracted main text

Abstract

The class of doubly-robust (DR) functionals studied by Rotnitzky et al. (2021) is of central importance in economics and biostatistics. It strictly includes both (i) the class of mean-square continuous functionals that can be written as an expectation of an affine functional of a conditional expectation studied by Chernozhukov et al. (2022b) and (ii) the class of functionals studied by Robins et al. (2008). The present state-of-the-art estimators for DR functionals $\psi$ are double-machine-learning (DML) estimators (Chernozhukov et al., 2018). A DML estimator $\widehat{\psi}_{1}$ of $\psi$ depends on estimates $\widehat{p} (x)$ and $\widehat{b} (x)$ of a pair of nuisance functions $p(x)$ and $b(x)$, and is said to satisfy "rate double-robustness" if the Cauchy--Schwarz upper bound of its bias is $o (n^{- 1/2})$. Were it achievable, our scientific goal would have been to construct valid, assumption-lean (i.e. no complexity-reducing assumptions on $b$ or $p$) tests of the validity of a nominal $(1 - \alpha)$ Wald confidence interval (CI) centered at $\widehat{\psi}_{1}$. But this would require a test of the bias to be $o (n^{-1/2})$, which can be shown not to exist. We therefore adopt the less ambitious goal of falsifying, when possible, an analyst's justification for her claim that the reported $(1 - \alpha)$ Wald CI is valid. In many instances, an analyst justifies her claim by imposing complexity-reducing assumptions on $b$ and $p$ to ensure "rate double-robustness". Here we exhibit valid, assumption-lean tests of $H_{0}$: "rate double-robustness holds", with non-trivial power against certain alternatives. If $H_{0}$ is rejected, we will have falsified her justification. However, no assumption-lean test of $H_{0}$, including ours, can be a consistent test. Thus, the failure of our test to reject is not meaningful evidence in favor of $H_{0}$.

Citation extraction

92
references
208
in-text mentions
93
distinct cited
27
self-citations
20,880
main-text words

appendix boundary found by appendix_command · 55% of the source is main text. Read the extracted text to check this.

Most heavily cited references

The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.

ReferenceIntensityMentionsSectionsMain text
1Liu, Lin, Mukherjee, Rajarshi, Robins, James M (2020) On nearly assumption-free tests of nominal confidence interval coverage for causal parameters estimated by machine learning self1.000114100%
2Liu, Lin, Mukherjee, Rajarshi, Newey, Whitney K, Robins, James M (2017) Semiparametric efficient empirical higher order influence function estimators self0.9285580%
3Ai, Chunrong, Chen, Xiaohong (2003) Efficient estimation of models with conditional moment restrictions containing unknown functions0.9285380%
4Rotnitzky, Andrea, Smucler, Ezequiel, Robins, James M (2021) Characterization of parameters with a mixed bias property self0.90916475%
5Chernozhukov, Victor, Chetverikov, Denis, Demirer, Mert, Duflo, Esth… (2018) Double/debiased machine learning for treatment and structural parameters self0.87482100%
6Newey, Whitney K, Robins, James M (2018) Cross-fitting and fast remainder rates for semiparametric estimation self0.87462100%
7Ai, Chunrong, Chen, Xiaohong (2007) Estimation of possibly misspecified semiparametric conditional moment restriction models with different conditioning variables0.8434375%
8Kline, Patrick, Saggio, Raffaele (2020) Leave-out estimation of variance components0.84333100%
9Robins, James, Li, Lingling, Tchetgen Tchetgen, Eric, Vaart, Aad (2008) Higher order influence functions and minimax estimation of nonlinear functionals self0.81413354%
10Chernozhukov, Victor, Newey, Whitney K, Singh, Rahul (2022) Automatic debiased machine learning of causal and structural effects0.81142100%

Showing the top 10 of 93 scored citations.

Cited by, within the corpus

arXiv econ.EM papers that cite this one, ranked by how heavily they lean on it.

Citing paperIntensityMentionsSections
1On the Asymptotic Inadmissibility of Double Machine Learning Estimators Under Structure-Agnostic Models1.00075
2Method-of-Moments Inference for GLMs and Doubly Robust Functionals under Proportional Asymptotics0.84343
3Stabilized Higher-Order Influence Functions: Statistical Theory of a Class of Bilinear Forms0.81142
4Higher-Order Debiased Estimators for General Treatment Models0.65973