Stability | experimental |
---|---|
Maintainer | Daniël de Kok <me@danieldk.eu> |
Safe Haskell | Safe-Inferred |
The more tests that are applied on a dataset, the more likely it becomes that a Type I error occurs (rejecting the null hypothesis when it is true). This module provides significance level corrections for repeated tests.
For more information, see:
- Bonferroni and Šidák corrections for multiple comparisons, H. Abdi, 2007, in: N.J. Salkind (ed.), Encyclopedia of Measurement and Statistics, Thousand Oaks, CA: Sage
- sidak :: (Floating f, Integral i) => f -> i -> f
- bonferroni :: (Fractional f, Integral i) => f -> i -> f
Documentation
:: (Floating f, Integral i) | |
=> f | Significance level |
-> i | Number of comparisons |
-> f | The corrected significance level |
Sidak correction: α* = 1 - (1 - α)^(/ n), where α is the intended significance level over all tests, n the number of tests, and α* the corrected significance level.
:: (Fractional f, Integral i) | |
=> f | Significance level |
-> i | Number of comparisons |
-> f | The corrected significance level |
Bonferroni correction: α* = / n, where α is the intended significance level over all tests, n the number of tests, and α* the corrected significance level.