Scribd
Upload
9 views3 pages

Mixture Methods for Multi-Exposure Analysis

The document provides an overview of mixture methods used for multi-exposure inference and prediction in environmental and biomedical research, focusing on challenges such as collinearity and confounding. It discusses the setup involving outcomes, exposures, and covariates, and outlines the targets for analysis including overall mixture effects and interactions. The goal is to enhance understanding of how various exposures impact outcomes in complex environments.

Uploaded by

190807wj
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
9 views3 pages

Mixture Methods for Multi-Exposure Analysis

The document provides an overview of mixture methods used for multi-exposure inference and prediction in environmental and biomedical research, focusing on challenges such as collinearity and confounding. It discusses the setup involving outcomes, exposures, and covariates, and outlines the targets for analysis including overall mixture effects and interactions. The goal is to enhance understanding of how various exposures impact outcomes in complex environments.

Uploaded by

190807wj
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd

Mixture Methods in Environmental

& Biomedical Research


BKMR · qgcomp · WQS · PCA · Elastic
Net · LASSO
Full 40-Minute Talk
Motivation & Context

• Goal: Overview of mixture methods for multi-exposure inference & prediction


• Context: Environmental/biomedical exposome, diet, omics, air pollution
• Challenges: Collinearity, nonlinearity, interactions, confounding
Notation & Setup

• Observations: outcome Y, exposures X1...Xp, covariates C


• Targets: overall mixture effect, partial effects, interactions, prediction
• Causal estimand example: E[Y | do(X=x)] - E[Y | do(X=x0)]

You might also like