I want to save the cross-sect specific coeffs from a pool estimation in series with the cross-sectional identifiers in the same way as the variables in my data are identified with cross-sectional identifiers. The following code does what I want:
Code: Select all
for !i = 1 to p_lrv99.@ncross
%ccy = p_lrv99.@idname(!i)
series c_lrv99_{%ccy} = c(37+!i)
next
Why do I want this? I'm trying to do a Fama-MacBeth type two-step asset-pricing in Eviews.
Step 1: time-series regression of each cross-section (asset) return on a common "risk factor"
Step 2: cross-section regression of "expected" return (say average) on coeffs from step 1.
Maybe a neater way to do this in Eviews?
Thanks in advance,
Dagfinn
