We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

stat.ML

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Statistics > Machine Learning

Title: A Finite Sample Theorem for Longitudinal Causal Inference with Machine Learning: Long Term, Dynamic, and Mediated Effects

Authors: Rahul Singh
Abstract: I construct and justify confidence intervals for longitudinal causal parameters estimated with machine learning. Longitudinal parameters include long term, dynamic, and mediated effects. I provide a nonasymptotic theorem for any longitudinal causal parameter in a general class, estimated with any machine learning algorithm that satisfies a few simple conditions. The main result encompasses local parameters defined for specific demographics as well as proximal parameters defined in the presence of unobserved confounding. I prove consistency, Gaussian approximation, and semiparametric efficiency. The rate of Gaussian approximation is $n^{-1/2}$ for global parameters, and it degrades gracefully for local parameters. I articulate a simple set of conditions to translate mean square rates into statistical inference, and verify that they hold for adversarial estimators over generic function spaces. A key feature of the main result is a new multiple robustness to ill posedness for proximal causal inference in longitudinal settings. Of independent interest, I provide what appears to be the first mean square rate for nested nonparametric instrumental variable regression.
Comments: 71 pages
Subjects: Machine Learning (stat.ML); Machine Learning (cs.LG); Econometrics (econ.EM); Statistics Theory (math.ST)
Cite as: arXiv:2112.14249 [stat.ML]
  (or arXiv:2112.14249v2 [stat.ML] for this version)

Submission history

From: Rahul Singh [view email]
[v1] Tue, 28 Dec 2021 18:29:56 GMT (44kb)
[v2] Mon, 3 Oct 2022 23:38:58 GMT (127kb)

Link back to: arXiv, form interface, contact.