In: Statistics and Probability
In the Benjamini-Hochberg procedure, is the critical value the same thing as the q-value? In other words, are the terms synonymous?
Yes, The critical value is the same thing as the q-value.
The Benjamini-Hochberg procedure is a powerful tool that reduces
the rate of false detection.
Speed control helps to control the fact that sometimes a small q
value (below 5%) occurs randomly, which may lead you to reject the
null hypothesis. In other words, the B-H procedure helps you avoid
Type I error (falsehood).
P-value of 5% means that there is only a 5% probability of getting
your results if the hypothesis is empty. In other words, if you get
a p -5% value, it is unlikely that your null hypothesis is invalid
and should be discarded. But this is just a probability - many
times the null hypothesis is simply thrown out because of the
randomness of the results.
In this case, let's say you have a group of 100 patients that you
know of no specific disease. Your meaningless hypothesis is that
the patients are free of disease and your choice is that they are
truly ill. If you run 100 statistical tests at a 5% alpha level,
then about 5% of the results would be considered positive.