Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: R: different output in different machines


From   "Carlo Lazzaro" <carlo.lazzaro@tin.it>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: R: different output in different machines
Date   Wed, 8 Dec 2010 13:45:44 +0100

Three basic questions following Jorge's question:
1) Are you sure you are using the same dataset and do-file in each case?
Something similar happened to me one week ago, but I was the only one to
blame because I have inadvertently overwritten the old version of my dataset
on the new .dta file. (Luckily enough I made some copies of the most recent
.dta file in advance!)
2) When you say that you're controlling for the version, do you mean that
Stata 10 is running on both the computers?
3) If yes, do both Stata 10 packages share the same updating (-help update
all-)?

Kind Regards,
Carlo
-----Messaggio originale-----
Da: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] Per conto di Prado De Castro
Alfaiate, Jorge
Inviato: mercoledì 8 dicembre 2010 13.18
A: statalist@hsphsun2.harvard.edu
Oggetto: st: different output in different machines

Dear all,

I have written a code that delivers a different output depending on the
computer in which I run it. I am controlling for the version, which is STATA
10. It is a mle, and in one computer I get a log-likelihood of, say,
1.176456, while in another computer I get, say, 1.1686789. That is, the
results are not radically different but they are different enough so that
they deliver significantly different coefficients in my regression. Also,
the speed of convergence changes (from, say, 25 iterations to, say, 32
iterations). I use the exact same dataset and do-file in each case...

Could anyone give me a clue about what may be causing this curious problem?

Thanks in advance,
Jorge

Jorge Prado de Castro Alfaiate
PhD student,
Department of Economics,
University of Essex

*
*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index