Statistical Methods for Environmental Mixtures and Exposome Research

EH 245 Subject & Catalog Number

Description

This course will cover statistical methods for the assessment of complex environmental exposures, such as pollutants and chemical mixtures, within the exposome framework. We will describe the analytical challenges involved in the joint evaluation of multiple environmental factors and provide an overview to a variety of analytical tools. Topics will include: data reduction techniques such as principal component analysis (PCA) and clustering; classical methods (multiple regression) and their limitations; variable selection techniques (Lasso, ridge regression, elastic net); statistical methods for environmental mixtures such as Bayesian Kernel Machine Regression (BKMR) and Weighted Quantile Sum (WQS) Regression; introduction to basic machine learning (ML) in environmental health; integrating environmental mixtures in health disparities research. The course will integrate lectures presenting the methods, case-studies from recently published papers, and hands-on data sessions. Lectures will present in a rigorous yet non-theoretical way the methods of interest, discussing when each method presented is appropriate for use and for which research question it can be applied.