A Neural Framework for Generalized Causal Sensitivity Analysis
This work addresses the problem of robust causal inference under unobserved confounding for researchers and practitioners, offering a flexible tool that is incremental in extending neural methods to sensitivity analysis.
The authors tackled the challenge of unobserved confounding in causal inference by proposing NeuralCSA, a neural framework for generalized causal sensitivity analysis that works with various sensitivity models, treatment types, and causal queries, achieving valid bounds as shown theoretically and empirically.
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) different treatment types (i.e., binary and continuous); and (iii) different causal queries, including (conditional) average treatment effects and simultaneous effects on multiple outcomes. The generality of NeuralCSA is achieved by learning a latent distribution shift that corresponds to a treatment intervention using two conditional normalizing flows. We provide theoretical guarantees that NeuralCSA is able to infer valid bounds on the causal query of interest and also demonstrate this empirically using both simulated and real-world data.