Extract intercepts from multiple regressions in stata

Question

I am attempting to reproduce the following in stata. This is a scatter plot of average portfolio returns (y axis) and predicted retruns (x axis).

To do so, I need your help on how I can extract the intercepts from 25 regressions into one variable? I am currently running the 25 portfolio regressions as follows. I have seen that parmest can potentially do this but can't get it to work with the forval. Many thanks

    forval s = 1 / 5 {
    forval h = 1 / 5 {
        reg S`s'H`h' Mkt_Rf SMB HML 

        }
    }

Answer 1

I don't know what your data look like, but maybe something like this will work:

gen intercepts = .
local i = 1
forval s = 1 / 5 {
    forval h = 1 / 5 {
        reg S`s'H`h' Mkt_Rf SMB HML 

        // assign the ith observation of intercepts
        // equal to the regression constant
        replace intercepts = _b[_cons] if _n == `i'

        // increment i
        local ++i
    }
}