|
Activity Number:
|
628
- Complex Data Analysis with Mental Health Applications
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 2, 2018 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Mental Health Statistics Section
|
|
Abstract #329331
|
Presentation
|
|
Title:
|
Estimation and Inference for the Mediation Effect in a Time-Varying Mediation Model
|
|
Author(s):
|
Donna Coffman* and Xizhen Cai and Runze Li
|
|
Companies:
|
Temple University and Temple University and Penn State University
|
|
Keywords:
|
mediation analysis;
time-varying effect;
intensive longitudinal data
|
|
Abstract:
|
Traditional mediation analysis studies the relationship between an intervention, a time-invariant mediator and a time-invariant outcome variable. The mediation effect from the treatment to the outcome through the mediator is usually assumed to be time-invariant. With the improvement of the technology, it is possible to make repeated assessments of subjects over time to obtain intensive longitudinal data. This calls for an extension of traditional mediation analysis to incorporate time-varying variables and effects. We consider a framework to build a time-varying mediation model, and focus on estimating and making inference regarding the time-varying mediation effect. We derive its asymptotic distribution at any fixed time point. We find the standard error formula and construct the corresponding point-wise confi dence band for the time-varying mediation effect. Simulation studies show good performance when comparing the con fidence band and the true underlying model. We conclude with a discussion of limitations and future extensions of the proposed method.
|
Authors who are presenting talks have a * after their name.
