https://github.com/nignatiadis/empirical-partially-bayes-paper

https://github.com/nignatiadis/empirical-partially-bayes-paper

Last synced: 26 days ago
JSON representation

• Host: GitHub
• URL: https://github.com/nignatiadis/empirical-partially-bayes-paper
• Owner: nignatiadis
• License: mit
• Created: 2022-12-03T03:04:21.000Z (about 2 years ago)
• Default Branch: main
• Last Pushed: 2024-04-18T03:51:03.000Z (9 months ago)
• Last Synced: 2024-04-18T20:04:34.123Z (9 months ago)
• Language: Julia
• Homepage: https://arxiv.org/abs/2303.02887
• Size: 14.8 MB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# empirical-partially-bayes-paper

Reproduction code for the paper:

> **Empirical partially Bayes multiple testing and compound χ² decisions.** Nikolaos Ignatiadis and Bodhisattva Sen, 2024, Annals of Statistics (forthcoming)

The repository structure is as follows:

* **data_analysis**: Folder that contains the three empirical analyses in the paper (methylation, microarray & proteomics).

* **simulations**: Folder for the reproduction of the simulation study of the paper.

* **average_significance_tightness.jl**: Example that average significance control inequality can be almost tight (mentioned in penultimate paragraph of Section 4.2).