multiple comparison correction

[jbao]

No description provided.

[lasomethingsomething]

Hey @jbao @mkolarek @elezar, Still planning on this? Want to add a Help Wanted label?

[jbao]

@LappleApple Yes, this is still relevant, some ideas from @robertmuil

• Decide on where to implement (I think in results is a good place)
• Decide when it will be performed (it will depend on the use, I think)
• Decide over which analyses it will be performed (I think within the 'type' is correct: all subgroups analysed in all SGAs, and, separately, all KPIs that have an uplift/delta calculation)
• Implement simes-hochberg with potential to select over which comparisons