A0254
Title: Inference for quantile mediation effects in the presence of complex confounding via deep neural networks
Authors: Shuoyang Wang - University of Louisville (United States) [presenting]
Yuan Huang - Yale University (United States)
Runze Li - The Pennsylvania State University (United States)
Abstract: Traditional mediation analysis methods face challenges when dealing with a large number of mediators. In practice, these challenges can be compounded by outliers and the complex relationships introduced by confounders. To address these issues, a novel quantile-based partially linear mediation analysis method is proposed that can handle high-dimensional mediators and introduce deep neural network techniques to model intricate relationships in confounders. Unlike most existing works that focus on mediator selection, inference on mediation effects is emphasized. Theoretical analysis shows that the proposed procedure controls type I error rates for hypothesis testing on mediation effects. When the dimension of the mediator is high, the proposed method consistently selects important features in the outcome model. Numerical studies show that the proposed method outperforms existing approaches under a variety of settings, demonstrating its versatility and reliability as a modeling tool for complex data. The application of the proposed method to study DNA methylation's mediation effect of childhood trauma on cortisol stress reactivity reveals previously undiscovered relationships by providing a comprehensive profile of the relationship at various quantiles.
