Great work!

Just to add on Mean-Variance section

I have worked several years with MV Optimizer to realize the following fact:
Most important input in MV optimization is Expected Return inputs E(r)

Ziemba & Chopra (1993) show that estimation error in Expected Return E(r)
is 10 times as important as estimation error in sigma (std-dev),
and 20 times as important as estimation error in rho (correlations)