Covariance matrix for two different models

[Bart Rokicki]

Is there a way to get covariance matrix of coefficients from two different models? I am trying to calculate the covariance for two coefficients from two different regressions.
Any help appreciated.

[Joseph Coveney]

Are you looking for something like this?

sysuse auto

logit foreign c.gear_ratio, nolog
estimates store G

logit foreign c.headroom, nolog
estimates store H

suest G H
matrix list e(V) // <- Here

[Bart Rokicki]

That would work - unfortunately I estimate instrumental variables model. Suest does not support it...

[Bart Rokicki]

But I bet there must be an easy way to do it.

[FernandoRios]

I think what you are looking for is wither sureg with "isure" option. Or reg3.