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 on April 23, and its replacement, statalist.org is already up and running.


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

st: Re: Same code, same machine, same data, different results


From   Mattia Landoni <mattia.landoni@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Re: Same code, same machine, same data, different results
Date   Thu, 6 Sep 2012 20:47:37 -0400

Thanks for the jokes and the constructive answers.

There is no m:m merge, and the same variables are dropped each time. The
problem was indeed one of sorting, as two of the answers mentioned,
together with a wrong use of indexing. There was a command such as
(simplifying)

sort id year
generate a_lag = a[_n - 1] if id == id[_n - 1]

the dataset is quarterly, so it has more than one observation per year.
Each time I ran the code, the dataset had a slightly different sorting and
some rows were assigned each time different values.

On Wed, Sep 5, 2012 at 5:10 PM, Mattia Landoni <mattia.landoni@gmail.com> wrote:
> Dear statalisters,
>
> a friend of mine has a bizarre problem. She is running a regression as follows:
>
> xi: regress a b c i.d i.e
>
> and her output is different every time. Has anyone ever seen a
> behavior like this? Below are some details.
>
> Environment:
> - Stata 11
> - Windows 32-bit
>
> Precise description:
> The do-file imports several files from .csv, then merges them, then
> runs the regression. If I run the do-file, I get certain results. If I
> issue the same regression command again, I get again the same results,
> as it should be. However, if I re-run the do-file from the beginning,
> I get slightly different results and the regression even reports a
> slightly different number of observations. (Say, 2663 vs. 2666). Every
> time all the data are taken afresh from the same static .csv sources.
> There is nothing random about the do-file, that I know. The xi:
> command generates about 200 i-variables and a few, maybe 10, are
> dropped because of collinearity. There are more than 2500
> observations.
>
> I could post the do-file here, but it's big and messy. If anyone has
> any insight after reading the above description, I'd be very glad to
> hear it.
>
> Thanks,
>
> Mattia
>
> --
> Mattia Landoni
> +1 217 4-A-QUANT (from US)
> +39 02 3206 21676 (from Italy)
> GMT -05:00 (US East)



-- 
Mattia Landoni
+1 217 4-A-QUANT (from US)
+39 02 3206 21676 (from Italy)
GMT -05:00 (US East)
*
*   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