B1421
Title: Statistical inference for high-dimensional generalized estimating equations
Authors: Lu Xia - Michigan State University (United States) [presenting]
Ali Shojaie - University of Washington (United States)
Abstract: A novel inference procedure is proposed for linear combinations of high-dimensional regression coefficients in generalized estimating equations (GEE), which are widely used to analyze correlated data. The estimator for this more general inferential target, obtained via constructing projected estimating equations, is shown to be asymptotically normally distributed under certain regularity conditions. A data-driven cross-validation procedure is also introduced to select the tuning parameter for estimating the projection direction, which is not addressed in the existing procedures. The robust finite-sample performance is demonstrated, especially in estimation bias and confidence interval coverage, of the proposed method via extensive simulations, and the method is applied to a longitudinal proteomic study of COVID-19 plasma samples to investigate the proteomic signatures associated with disease severity.
