Because -version- is not a time machine. -version- sets the clock back
in terms of behaviour that you might have good reason for wanting, and
StataCorp agrees with you; it doesn't reverse bug fixes or what
StataCorp regards as improvements.
The help at -version- tells you exactly what -version- does, and the
executive summary is
(a) exactly what it says there, and no more
(b) so, less than you might think
There is a fuller story at
http://www.stata.com/support/faqs/programming/user-written-programs-and-stata-version/
The question in that FAQ is not your question, but that is a more
discursive account which might be helpful.
Otherwise put, if you want exactly what 9.2 did, using Stata 9.2 is
the only way to guarantee that.
Nick
njcoxstata@gmail.com
On 12 April 2013 11:19, Kamira Sanchez Nicosia
<kamira.sanchez@uni-jena.de> wrote:
> I'm using Stata 11 and even if I set the version 9.2 the standard errors with areg and xtreg are computed as using version 10 or greater. Does anyone know why?
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/
