P-value correction: Difference between revisions
Content deleted Content added
reinstated the TASK's name and category (it was deleted when promoting this draft task). |
m added whitespace to the task's preamble, add whitespace before the table of contents. |
||
Line 1:
{{task|Probability and statistics}}
Given a list of [[wp:p-value|p-values]], adjust the p-values for multiple comparisons. This is done in order to control the false positive, or Type 1 error rate.
This is also known as the "[[wp:False discovery rate|false discovery rate]]" (FDR). After adjustment, the p-values will be higher but still inside [0,1].
The adjusted p-values are sometimes called "q-values".
;Task:
Line 16 ⟶ 21:
1.801145e-05, 2.504456e-07, 3.310253e-02, 9.427839e-03, 8.791153e-04,
2.177831e-04, 9.693054e-04, 6.610250e-05, 2.900813e-02, 5.735490e-03}
There are several methods to do this, see:
Line 23 ⟶ 29:
* Yosef Hochberg, "A sharper Bonferroni procedure for multiple tests of significance", ''Biometrika'', Vol. 75, No. 4 (1988), pp 800–802, DOI:[https://doi.org/10.1093/biomet/75.4.800 10.1093/biomet/75.4.800] JSTOR:[https://www.jstor.org/stable/2336325 2336325]
* Gerhard Hommel, "A stagewise rejective multiple test procedure based on a modified Bonferroni test", ''Biometrika'', Vol. 75, No. 2 (1988), pp 383–386, DOI:[https://doi.org/10.1093/biomet/75.2.383 10.1093/biomet/75.2.383] JSTOR:[https://www.jstor.org/stable/2336190 2336190]
Each method has its own advantages and disadvantages.
<br><br>
=={{header|C}}==
| |||