Photo: ayeletb, CC BY-SA 1.0, Wikimedia Commons
Yoav Benjamini
Tel Aviv University
- Research field
- Multiple comparisons, selective inference, the false discovery rate
Yoav Benjamini and Yosef Hochberg proposed in 1995 to replace the question of whether any false discovery has been made by the question of what fraction of the discoveries are false, defined that fraction's expectation as the false discovery rate, and gave a simple rule on the sorted p-values that keeps it below a chosen level for independent tests. The procedure became the standard of genomics and brain imaging within a decade, and his later work extended the guarantee to dependent tests. The unit works their own example, the fifteen p-values of a heart attack trial, and uses the rule on a simulated screen of 10,000 genes.
In the masterclass
Section 2, after the Benjamini-Hochberg procedure, its guarantee, and the worked example on the fifteen p-values of the trial that the paper itself used.
The research is presented in Unit 16: Multiple Testing and the False Discovery Rate, in a research spotlight box that explains the result at the level of the unit and links it to what you have just learned.