Copula-based Sensitivity Analysis for Multi-Treatment Causal Inference with Unobserved Confounding

A Kernel Test for Causal Association via Noise Contrastive Backdoor Adjustment

Interpreting population- and family-based genome-wide association studies in the presence of confounding

GWASs aim to estimate direct effects of genotype on an individual’s phenotype, but this can be subject to genetic and environmental confounds and "indirect" genetic effects of relatives’ genotypes. This study presents a theoretical analysis of the influence of confounders in population-based and within-family GWASs, showing that, while family-based studies are more rigorous, they still carry subtle issues that arise from confounding.

High-Dimensional Inference for Generalized Linear Models with Hidden Confounding

Scalable Computation of Causal Bounds

The Proximal ID Algorithm

Naive regression requires weaker assumptions than factor models to adjust for multiple cause confounding