Print  Close

Individualized Dynamic Mediation Analysis Using Latent Factor Models

Abstract Number:

3268 

Submission Type:

Contributed Abstract 

Contributed Abstract Type:

Paper 

Participants:

Yijiao Zhang (1), Yubai Yuan (2), Yuexia Zhang (3), Zhongyi Zhu (4), Annie Qu (5)

Institutions:

(1) Fudan University, Shanghai, CHNIA, (2) Pennsylvania State University, N/A, (3) The University of Texas at San Antonio, N/A, (4) Fudan University, N/A, (5) University of California At Irvine, Irvine, CA

Co-Author(s):

Yubai Yuan  
Pennsylvania State University
Yuexia Zhang  
The University of Texas at San Antonio
Zhongyi Zhu  
Fudan University
Annie Qu  
University of California At Irvine

First Author:

Yijiao Zhang  
Fudan University

Presenting Author:

Yijiao Zhang  
N/A

Abstract Text:

Mediation analysis plays a crucial role in causal inference as it can investigate the pathways through which the treatment influences the outcome. However, mediation effects usually unfold over time and exhibit significant heterogeneity within populations. Additionally, the presence of unobserved confounding variables poses a significant challenge in causal inference. To address these issues, we propose an individualized dynamic mediation analysis method based on factor models. A key advantage of our method is that we can recover individualized dynamic effects in the presence of unmeasured time-varying confounders. Besides, our method also enables simultaneous clustering of both individuals and mediators to further improve the estimation efficiency. We provide estimation consistency as well asymptotic normality for our proposed estimators. Extensive simulation studies and an application to DNA methylation study demonstrate the effectiveness and superiority of our method.

Keywords:

Mediation analysis|Heterogeneous effects|Unmeasured confounding|Factor model|Structural equation model|Variable selection

Sponsors:

Biometrics Section

Tracks:

Personalized/Precision Medicine

Can this be considered for alternate subtype?

Yes

Are you interested in volunteering to serve as a session chair?

No

I have read and understand that JSM participants must abide by the Participant Guidelines.

Yes

I understand that JSM participants must register and pay the appropriate registration fee by June 1, 2024. The registration fee is non-refundable.

I understand