conditional_value_at_risk seems to be wrong

[Andrei-Oluin]

The function conditional_value_at_risk in stats seems to have two errors.

• The way c_var is calculated seems to always make it NaN and thus the function always returns var
• Shouldn't it be returns[returns > var] instead of returns[returns < var]? Now this might be entirely wrong - I'm just looking at the german wiki https://de.wikipedia.org/wiki/Conditional_Value_at_Risk which states $CVaR_{\alpha }(X)= E [X|X> VaR _{\alpha }(X)]$ citing https://sci-hub.st/https://doi.org/10.1111/1467-9965.00068.