Combining Independent, Weighted P-Values: Achieving Computational Stability by a Systematic Expansion with Controllable Accuracy
2011

Combining Weighted P-Values for Stability

publication Evidence: high

Author Information

Author(s): Alves Gelio, Yu Yi-Kuo

Primary Institution: National Center for Biotechnology Information, National Library of Medicine, National Institutes of Health

Hypothesis

Can a systematic expansion method be developed to combine weighted p-values while maintaining computational stability?

Conclusion

The study presents a new method for combining weighted p-values that ensures both accurate statistics and stable computations, even when weights are nearly degenerate.

Supporting Evidence

  • The new method provides a systematic way to combine p-values.
  • It addresses the issue of numerical instability when weights are nearly degenerate.
  • The approach is applicable across various scientific disciplines.

Takeaway

This study shows a way to combine different statistical tests into one result without making mistakes, even when the tests are very similar.

Methodology

The authors derived a controlled expansion for combining p-values from different groups, ensuring numerical stability and accuracy.

Limitations

The method assumes that the p-values being combined are independent, which may not always be the case in real applications.

Digital Object Identifier (DOI)

10.1371/journal.pone.0022647

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication