Re: Regression: conditional CAPM with indicator as an instrument
Posted: Sun Jul 23, 2017 2:19 pm
I don't understand where the numbers are from.
My guess would be the straightforward version, although obviously I don't know what the original authors did.The numbers are from regression (2) then applied in regression (3), as I see in the first bracket of equation for regression 3: [α0 + αB · IB,t−1] so α0 is the C and αB · IB,t−1 is the alpha from regression (2) 0.024167*i(-1).
If that's not the case then my estimation for regression 3 would be :
ls rw c i(-1) rm i(-1)*rm i(-1)*x*rm . This seems too straight forward, but then again it might be the right equation...
What would you say?