Hidden yet quantifiable: A lower bound for confounding strength using randomized trials

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: De Bartolomeis, Piersilvio, Abad, Javier, Donhauser, Konstantin, Yang, Fanny
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914407251968000
author De Bartolomeis, Piersilvio
Abad, Javier
Donhauser, Konstantin
Yang, Fanny
author_facet De Bartolomeis, Piersilvio
Abad, Javier
Donhauser, Konstantin
Yang, Fanny
contents In the era of fast-paced precision medicine, observational studies play a major role in properly evaluating new treatments in clinical practice. Yet, unobserved confounding can significantly compromise causal conclusions drawn from non-randomized data. We propose a novel strategy that leverages randomized trials to quantify unobserved confounding. First, we design a statistical test to detect unobserved confounding with strength above a given threshold. Then, we use the test to estimate an asymptotically valid lower bound on the unobserved confounding strength. We evaluate the power and validity of our statistical test on several synthetic and semi-synthetic datasets. Further, we show how our lower bound can correctly identify the absence and presence of unobserved confounding in a real-world setting.
format Preprint
id arxiv_https___arxiv_org_abs_2312_03871
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Hidden yet quantifiable: A lower bound for confounding strength using randomized trials
De Bartolomeis, Piersilvio
Abad, Javier
Donhauser, Konstantin
Yang, Fanny
Machine Learning
In the era of fast-paced precision medicine, observational studies play a major role in properly evaluating new treatments in clinical practice. Yet, unobserved confounding can significantly compromise causal conclusions drawn from non-randomized data. We propose a novel strategy that leverages randomized trials to quantify unobserved confounding. First, we design a statistical test to detect unobserved confounding with strength above a given threshold. Then, we use the test to estimate an asymptotically valid lower bound on the unobserved confounding strength. We evaluate the power and validity of our statistical test on several synthetic and semi-synthetic datasets. Further, we show how our lower bound can correctly identify the absence and presence of unobserved confounding in a real-world setting.
title Hidden yet quantifiable: A lower bound for confounding strength using randomized trials
topic Machine Learning
url https://arxiv.org/abs/2312.03871