Abhinandan Dalal, Patrick Blöbaum, Shiva Kasiviswanathan, Aaditya Ramdas
arXiv 18 Aug 2024 · Statistics — Methodology
arXiv:2408.09598 · PDF · DOI · OpenAlex · Extracted main text
Double (debiased) machine learning (DML) has seen widespread use in recent years for learning causal/structural parameters, in part due to its flexibility and adaptability to high-dimensional nuisance functions as well as its ability to avoid bias from regularization or overfitting. However, the classic double-debiased framework is only valid asymptotically for a predetermined sample size, thus lacking the flexibility of collecting more data if sharper inference is needed, or stopping data collection early if useful inferences can be made earlier than expected. This can be of particular concern in large scale experimental studies with huge financial costs or human lives at stake, as well as in observational studies where the length of confidence of intervals do not shrink to zero even with increasing sample size due to partial identifiability of a structural parameter. In this paper, we present time-uniform counterparts to the asymptotic DML results, enabling valid inference and confidence intervals for structural parameters to be constructed at any arbitrary (possibly data-dependent) stopping time. We provide conditions which are only slightly stronger than the standard DML conditions, but offer the stronger guarantee for anytime-valid inference. This facilitates the transformation of any existing DML method to provide anytime-valid guarantees with minimal modifications, making it highly adaptable and easy to use. We illustrate our procedure using two instances: a) local average treatment effect in online experiments with non-compliance, and b) partial identification of average treatment effect in observational studies with potential unmeasured confounding.
appendix boundary found by appendix_command · 80% of the source is main text. Read the extracted text to check this.
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.
| Reference | Intensity | Mentions | Sections | Main text | |
|---|---|---|---|---|---|
| 1 | Victor Chernozhukov, Denis Chetverikov, Mert Demirer, Esther Duflo,… (2018) Double/debiased machine learning for treatment and structural parameters | 0.964 | 19 | 6 | 89% |
| 2 | Ian Waudby-Smith, David Arbour, Ritwik Sinha, Edward H Kennedy, and… (2024) Time-uniform central limit theory and asymptotic confidence sequences self | 0.933 | 16 | 7 | 81% |
| 3 | Yachong Yang, Arun Kumar Kuchibhotla, and Eric Tchetgen Tchetgen (2023) Forster-Warmuth counterfactual regression: A unified learning approach | 0.928 | 4 | 3 | 100% |
| 4 | Steven R. Howard, Aaditya Ramdas, Jon McAuliffe, and Jasjeet Sekhon (2021) Time-uniform, nonparametric, nonasymptotic confidence sequences self | 0.874 | 5 | 2 | 100% |
| 5 | Jerzy Neyman (1959) Optimal asymptotic tests of composite hypotheses | 0.811 | 4 | 2 | 100% |
| 6 | Edward H Kennedy, Zongming Ma, Matthew D McHugh, and Dylan S Small (2017) Non-parametric methods for doubly robust estimation of continuous treatment effects | 0.737 | 3 | 2 | 100% |
| 7 | Whitney K Newey (1994) The asymptotic variance of semiparametric estimators | 0.737 | 3 | 2 | 100% |
| 8 | Steve Yadlowsky, Hongseok Namkoong, Sanjay Basu, John Duchi, and Lu… (2022) Bounds on the conditional and average treatment effect with unobserved confounding factors | 0.669 | 10 | 3 | 30% |
| 9 | Peter J Bickel, Chris AJ Klaassen, Peter J Bickel, Ya’acov Ritov, J… (1993) Efficient and adaptive estimation for semiparametric models, volume 4 of Johns Hopkins Series in the Mathematical Sciences | 0.644 | 2 | 2 | 100% |
| 10 | Victor Chernozhukov, Juan Carlos Escanciano, Hidehiko Ichimura, Whit… (2022) Locally robust semiparametric estimation | 0.644 | 2 | 2 | 100% |
Showing the top 10 of 101 scored citations.
arXiv econ.EM papers that cite this one, ranked by how heavily they lean on it.
| Citing paper | Intensity | Mentions | Sections | |
|---|---|---|---|---|
| 1 | Partial Identification of Causal Effects for Endogenous Continuous Treatments | 0.405 | 1 | 1 |