Causal Discovery with Generalized Linear Models through Peeling Algorithms
Author Information
Author(s): Wang Minjie, Shen Xiaotong, Pan Wei
Primary Institution: Binghamton University, State University of New York
Hypothesis
Can a novel method for causal discovery using generalized linear models effectively identify causal relationships in the presence of unmeasured confounders?
Conclusion
The proposed GAMPI method successfully identifies causal relationships and valid instruments, demonstrating superior performance compared to existing methods.
Supporting Evidence
- The GAMPI method effectively reconstructs causal networks involving gene-to-gene and gene-to-disease relationships.
- Numerical experiments show GAMPI outperforms state-of-the-art methods in identifying causal relationships.
- The method is applicable to various data types, including discrete and continuous outcomes.
Takeaway
This study introduces a new way to find cause-and-effect relationships between variables, even when some important information is missing.
Methodology
The study employs a two-step process involving a fidelity model and peeling algorithms to identify causal relationships and correct for confounding effects.
Potential Biases
Potential biases may arise from unmeasured confounders that are not accounted for in the model.
Limitations
The method's performance may vary depending on the underlying assumptions about the data and the presence of confounders.
Participant Demographics
The study analyzed data from 712 subjects, including individuals with Alzheimer's disease and healthy controls.
Statistical Information
P-Value
p<0.05
Statistical Significance
p<0.05
Want to read the original?
Access the complete publication on the publisher's website