|
Activity Number:
|
69
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, July 31, 2016 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Biometrics Section
|
|
Abstract #320613
|
|
|
Title:
|
Nonlinear Structural Equation Models in Environmental Epidemiology
|
|
Author(s):
|
Esben Budtz-Jørgensen*
|
|
Companies:
|
|
|
Keywords:
|
latent variable ;
non-linear ;
robust ;
mis-specification
|
|
Abstract:
|
Structural equation models (SEMs) have been shown to be powerful for drawing causal inference in high dimensional data. In the measurement part of the model, observed variables are linked to a limited number of latent variables which are then related to each other in a structural model. The classical SEM builds on strong linearity assumptions. This is a serious limitation in applications which will often require in the introduction of non-linear terms. This paper considers methods for estimating and identifying non-linear associations in the SEM framework. Methods are illustrated in epidemiological data on health effects of prenatal mercury exposure.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2016 program
|
