Estimating the Proportion of True Null Hypotheses for Multiple Comparisons
Author Information
Author(s): Jiang Hongmei, Doerge R.W.
Primary Institution: Department of Statistics, Northwestern University
Hypothesis
How can we accurately estimate the proportion of true null hypotheses in multiple comparison situations?
Conclusion
The proposed method for estimating the proportion of true null hypotheses shows relatively small bias and variance compared to existing methods.
Supporting Evidence
- The proposed method provides estimates with relatively small bias and variance.
- Simulation studies show that the average estimate approach outperforms existing methods.
- The method is easy to implement and can be automated.
Takeaway
This study helps scientists figure out how many of their tests are actually showing real results instead of just random chance.
Methodology
The study proposes a simple method for estimating the proportion of true null hypotheses using simulated and real data comparisons.
Potential Biases
There is a risk of upward bias when estimating the proportion of true null hypotheses.
Limitations
The method may still have biases and the performance can vary based on the chosen parameters.
Want to read the original?
Access the complete publication on the publisher's website