Entering edit mode
8.6 years ago
ckhh
•
0
Hi,
I'm trying to use 'qvalue' package in r to work out the true proportion of alternatives.
I get pi0 is >1 with small lambda. pi0.lambda
seems to decrease as lambda increases.
In my simulated dataset I have only a very small proportion of true alternatives. (approx 0.001%)
Can this be the reason? Is there any way I can work out the true proportion of alternative test?