Simulation-based FDR estimation

[gokceneraslan]

Especially for genomics applications, it would be nice to also have FDR procedure in the implementation.

[hofnerb]

Thanks. This is new to me. I will have a look at it (but not in a timely manner). If you like, you may contribute relevant code via a pull request.

Btw. please note that we do not controll the FWER (P(V >= 1)) but the PFER (E(V)) which is far less conservative and may be >> 1. As the PFER controls (very, very conservatively) the FWER, stability selection can also be used to control the latter but usually is not!