stepwise estimation of seemingly unrelated regressions (SUR) - help

[Roni Golan]

Hello there,

Do any of you have an idea how can I do a stepwise estimation with seemingly unrelated regressions?

I have tried to do it in two steps: 1) stepwise regressions 2) use the SUR procedure

later I have tried to switch the order 1) SUR procedure over the full model 2) manually drop insignificant independent variable and go back to 1...

which one makes more sense? is there a better way?

Any response would be appreciated,

Roni

[Maarten Buis]

The typical advise on this forum is to not use stepwise at all: http://www.stata.com/support/faqs/st...sion-problems/

[Roni Golan]

Thanks!