Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

Re: st: Debugging estimation changes


From   Nick Sanders <[email protected]>
To   "[email protected]" <[email protected]>
Subject   Re: st: Debugging estimation changes
Date   Fri, 22 Jun 2012 13:38:32 -0700

Hi Ed,

My first guess would be a seed issue - do you have a sample or similar (e.g., psmatch2) command in your code? If so, did you set a seed?

Just a thought,
Nick 



On Jun 22, 2012, at 1:16 PM, Ed Gerrish <[email protected]> wrote:

> This is a general question about debugging STATA changes.  Here is my dilemma:
> 
> About 8 months ago, I ran a set of models and got results.  I saved
> them as my baseline results, wrote the paper, and worked on
> robustness.  Recently, I noticed that results that should have been
> similar to the baseline were not.  When I went back to run the
> baseline, the results had changed.  I figured I made a coding error
> and I did the following:
> 
> 1. Checked my code for recent changes.
> 2. Uncompressed my archived do files and datasets and ran them - the
> new results persisted.
> 3. Checked to make sure version control was on (it was).
> 4. Checked my custom ado's for updates.  There were some, but when run
> on older versions, the new results were sustained.
> 
> Any ideas on how I should proceed?  It is unclear to me what my next
> step is.  This certainly could be a STATA 11 vs. 12 problem -- but I
> was hoping version control would prevent that from being true.
> 
> Thanks,
> 
> -Ed
> *
> *   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 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index