Minimize C_p

This is also known as : Mallow's C_p-criterion. In this model, the utility function is given by:

C_p - n = (m - n_t) * Var(e) + (n - m) * Var(e_t)

where

Var(e_t) = variance of excess return of manager over benchmark, using all indices
Var(e) = variance of excess return of manager over benchmark, using current subset
n = number of indices in current subset
n_t = total number of indices
m = number of returns

Mallow's criterion calls for minimizing the utility function, i.e., StyleADVISOR chooses the subset of indices where C_p - n assumes its minimal value.

Related Statistics:
Model Selection
Maximize Adjusted R2
Minimize SC
Minimize AIC