st: RE: Spooky: Different Results with Repeated FE Estimation

["Sarah Edgington" <sedging@ucla.edu>]

Miyu,
Do you get different results if you run the same dofile repeatedly?  In
situations where I've run into issues with results changing across multiple
runs of the same dofile it has almost always turned out to be an issue where
the sort order affects something that I am doing.   Usually this is because
I'm doing some calculation that is dependent on the ordering of records but
I haven't actually sorted on a unique set of identifiers first.  So that's
one of the first things I'd double check.

If you run the models multiple times in the same dofile and get different
results across the different models, but they're the same every time you run
the dofile (i.e. you get the same results each instance of the model each
time you run the dofile but the first and second instance of the model in
the dofile differ) then it's more likely that the problem is that you've
changed either the data or the model in some subtle way.
Good luck.
-Sarah

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Miyu Lee
Sent: Wednesday, April 25, 2012 9:19 AM
To: statalist@hsphsun2.harvard.edu
Subject: st: Spooky: Different Results with Repeated FE Estimation

Hello everyone,

I was wondering whether anyone has made a similar experience in the
past: I am running a FE estimation with the SAME data and the SAME model but
for some reason I get different results every time I run it.

I do not have any "special" commands in my do file except some basic
mathematical operations. Could it be because I also have some log
transformations?

It is just really strange to me.

Thanks in advance.

Miyu
*
*   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/