Can anyone provide a more mathematically rigorous derivation of the factor model covariance formula. I’m just trying to understand how this is constructed better to help remember it.
Cov(i,j)=Bi1*Bj1*Var(F1)+Bi2*Bj2*Var(F2)+(Bi1*Bj2+Bi2*Bj1)*Cov(F1,F2)
Any ideas as to how this is derived?
It is multi factor formula derived from portfolio variance formula. I will concentrate to learn shorter version based only on one factor:
cov(i, j) = BiBj*VARm
Just came up with the same question…
Update…
You should know both versions.
Good luck!
Update: we have subscripts.
We should use them.