A Neural Framework for Generalized Causal Sensitivity Analysis

arXiv:2311.16026v2 Announce Type: replace
Abstract: Unobserved confounding is common in many applications, making causal inference from observational data challenging. As a remedy, causal sensitivity analysis is an important tool to draw causal conclusions under unobserved confounding with mathematical guarantees. In this paper, we propose NeuralCSA, a neural framework for generalized causal sensitivity analysis. Unlike previous work, our framework is compatible with (i) a large class of sensitivity models, including the marginal sensitivity model, f-sensitivity models, and Rosenbaum's sensitivity model; (ii) …

abstract analysis applications arxiv causal causal inference confounding cs.lg data framework generalized inference making paper sensitivity stat.ml tool type