Hi everyone, I recently ran into a problem while trying out the policy efficient frontier (pictured). Different policy rules generate huge differences in the volatility of the target variable, making it seem impossible to compare policies. I would like to ask:
(1) Is this normal?
(2) How to adjust the method to change this state?
My loss function is set as min{omega*var(x1)+(1-omega)*var(x2)}, x1,x2∈[Y, B, E]
