Online Program Home
My Program

Abstract Details

Activity Number: 464 - SPEED: Infectious Diseases, Spatial Modeling and Environmental Exposures, Speed 1
Type: Contributed
Date/Time: Wednesday, July 31, 2019 : 8:30 AM to 10:20 AM
Sponsor: Section on Statistics in Epidemiology
Abstract #304837
Title: Evaluation of Semiparametric Single Index Model for Characterizing Effects of Correlated Exposures
Author(s): Yuyan Wang* and Mengling Liu
Companies: New York University and New York University
Keywords: Correlation; Exposure mixture; Simulation; Single index model; Weighted quantile sum regression

Characterization of effects of correlated exposure mixture on an adverse outcome remains a challenging problem in environmental epidemiology. These multiple environmental exposures occur simultaneously, often have a complex correlation structure and display interplay effects on the outcome. Semiparametric single index model (SIM) is an appealing statistical method that enables the estimation of overall effect of multiple exposure mixtures, with the index coefficients representing the direction and weight from each component. We thus evaluate the performance of semiparametric SIM in characterizing effects and directions of correlated exposure mixture under various settings of correlation structure patterns and magnitudes of effect sizes. In addition, we show that the semiparametric SIM can overcome the limitation of the weighted quantile sum (WQS) regression, which requires the effects of all included exposures to be in the same direction. Extensive simulations and a real data application will be presented.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program