Re: st: Two same dataset but results says problem of collinearity

[Sergiy Radyakin <serjradyakin@gmail.com>]

On Thu, Jun 30, 2011 at 12:33 PM, vikramfinavker
<vikramfinavker@gmail.com> wrote:
> Dear All,
>
> I have two identical datasets. However, when i run regression using one
> dataset it says two of my year dummies are dropped due to the problem of
> collinearity but when i run same regression using other database(same) it
> drops all the year dummies and says they are collinear. It will be really
> very helpful to me if anyone of you could suggest me whats going on here?

Potential answers:

1) datasets are not the same; --> compute checksums to verify;
2) regressions are not the same; --> look at trace immediately before
estimation;
3) path dependency --> result depends on the internal state of the
optimizer. If you are running two estimations in one session you may
want to try set sortseed.
4) computers are not the same; --> precision-related problems;
different Stata versions, etc
... plenty of other explanations, writing more informative questions
will allow more informative answers. 4 is extremely rare as Stata
developers put every effort to ensure reproducibility of results.

Best, Sergiy Radyakin

>
> Regards,
>
> Vikram
>
> --
> View this message in context: http://statalist.1588530.n2.nabble.com/Two-same-dataset-but-results-says-problem-of-collinearity-tp6534025p6534025.html
> Sent from the Statalist mailing list archive at Nabble.com.
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
>

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/