Module multiple_testing 

Multiple comparison correction methods

When performing many simultaneous hypothesis tests, the probability of at least one false positive (Type I error) increases. Multiple testing corrections adjust p-values or significance thresholds to control for this.

Methods Provided

• Bonferroni - Controls the family-wise error rate (FWER) by multiplying each p-value by m
• Holm-Bonferroni (step-down) - A more powerful step-down variant of Bonferroni
• Hochberg (step-up) - A step-up variant that is more powerful than Holm under independence
• Benjamini-Hochberg - Controls the false discovery rate (FDR)
• Benjamini-Yekutieli - Controls FDR under arbitrary dependence
• Sidak - Uses the Sidak correction: 1 - (1 - alpha)^(1/m)

Structs

MultipleCorrectionResult

Result of a multiple testing correction

Functions

benjamini_hochberg

Applies the Benjamini-Hochberg procedure for FDR control.

benjamini_yekutieli

Applies the Benjamini-Yekutieli procedure for FDR control under arbitrary dependence.

bonferroni

Applies Bonferroni correction to a set of p-values.

hochberg

Applies the Hochberg step-up correction.

holm_bonferroni

Applies the Holm-Bonferroni step-down correction.

multipletests

Applies a multiple testing correction by method name.

sidak

Applies the Sidak correction.