----- Original Message -----
From: <[email protected]>
To: <[email protected]>
Sent: Monday, October 13, 2003 7:01 AM
Subject: st: SUR and Panels
> Dear Statalisters,
>
> I have the following data format:
>
> i(1,2,3..N) t var1 var2 var3
> 1 91 . . .
> 1 92 . . .
> 1 93 . . .
> . .
> 2 91 . .
> 2 92 . .
> 2 93 . .
> . .
> 3 91 . .
> 3 92 . .
> 3 93 . .
> . .
> 4 91 .
> . . .
>
>
> Let i(1,2,3,...,N) be cross-sections and t the time operator. I want to run
> Seemingly Unrelated Regressions (sureg), for one equation per cross section,
> ie:
>
> sureg (1_var1 1_var2 1_var3) (2_var1 2_var2 3_var3) ... (N_var1 N_var2 Nvar3)
>
> I didnt do my homework on panel data analysis in STATA, but I do know that the
> panels are balanced. Every suggestion is much appreciated,
> many thanx,
> Dimitris
>
In order to run -sureg- you need to reshape your data into the wide format (see
[R] reshape).
I believe the following should work (taken from an example by Kit Baum
10/22/02):
reshape wide var1 var2 var3, i(t) j(i)
forv i=1/N {
local rhs "`rhs' ( `i'_var1 `i'_var2 `i'_var3) "
}
di "`rhs'"
sureg `rhs'
Hope this helps,
Scott
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
