Hi there,
I am trying to do a Fama-MacBeth style regression. I have read through the very helpful example that was downloaded with the add-in.
In that there are a number of variables pr11 etc. and a number of factors, however, the factors are the same for all variables through time. What I want to do is regress asset returns on factors that are different for each firm, i.e. regress each firm's return on its own beta (for that month), its own market cap (for that month) and a dummy for whether it has foreign sales or not, and each of these factors can vary monthly over 20 years. I can see how I could step 1 manually for each firm and get coefficients for each of these variables, but then how would I conduct the second step in Eviews? Could I just overwrite the beta estimates in bhat01? Or could Eviews do this for me, using a matrix for each factor for each firm? Thanks for your help, Martha.
Fama-MacBeth Add-In
Moderators: EViews Gareth, EViews Moderator, EViews Esther
Who is online
Users browsing this forum: No registered users and 35 guests