Unraveling time-varying causal effects of multiple exposures: integrating Functional Data Analysis with Multivariable Mendelian Randomization
Published:Dec 22, 2025 06:06
•1 min read
•ArXiv
Analysis
This article proposes a novel methodology by combining Functional Data Analysis (FDA) with Multivariable Mendelian Randomization (MR) to investigate time-varying causal effects of multiple exposures. The integration of these two methods is a significant contribution, potentially allowing for a more nuanced understanding of complex causal relationships in various fields. The use of FDA allows for the modeling of exposures and outcomes as continuous functions over time, while MR leverages genetic variants to infer causal relationships. The combination offers a powerful approach to address the limitations of traditional MR methods when dealing with time-varying exposures. The article's focus on integrating these methodologies suggests a focus on methodological advancement rather than a specific application or result.
Key Takeaways
Reference
“The article focuses on methodological advancement by integrating FDA and MR.”