Testing preference parameters in choice models

[mar vas]

Dear all,

I am trying to test parameters between two models of different demographic.
so my hypothesis is
h1: beta1= beta 2
h2: beta1<>beta2

I estimate both models and note the values of beta1 and beta 2 and the difference between them
Lets say is x= beta1-beta2

now the Standard error of (x) is calculated as
se(x)= sqrt[var(beta1) +var(beta2)- 2 covariance(beta1,beta2)]

1. var i.e. the variance I find using the variance-covariance matrix but don't see how to use the covariance in this formula.

2. I was computing this on excel but perhaps there is a quicker way through stata?

Any help would be appreciated.

Thank you
Mar