In simple linear regression, the beta is b1: y = b0 + b1X In multiple linear regression, the equation is: y = b0 + b1X1 + b2X2 + b3x3. So which coefficient (b1, b2, or b3–if any) is the beta?
Typically it would be b1, but I don’t think it makes a difference, your answer will still be the same.
B1, B2, and B3 are all betas based upon different independent variables and b0 is the Y-interecpt.
right. i think b1 is the beta assuming b2 and b3 are zero, b2 is the beta assuming b1 and b3 are zero, and b3 is the beta assuming b1 and b2 are zero. dont think they would ask you a Q where it didnt say holding everything else at zero or something like that. thanks guys–nice to see you back in action bpuldog.
beta typically is associated with the Market Premium ( Which is usually the first variable ) , while the others might be HML or SML (high PB minus Low PB and small cap minus large cap )
There is no beta for the regression, only individual factor betas. b1, b2, and b3 are all betas even if they are 0. A beta of 0 means the dep. variable is not sensitive to that factor. Just b/c the indep. variable doesn’t contribute to the dep. variable doesn’t mean it’s coefficient stops being a beta. even in a simple regression there is no beta for the regression only a beta for the independent variable.
Beta is a vector, so it is all b’s coefficients to the X matrix of M independant variables. With proper notation you may see Beta=(1,1,5,2,14) for an M=5 regression.
B1, B2, and B3 are all betas based upon different independent variables and b0 is the Y-interecpt.
Sorry for reposting my bad.