Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Nick Cox <n.j.cox@durham.ac.uk> |
To | "'statalist@hsphsun2.harvard.edu'" <statalist@hsphsun2.harvard.edu> |
Subject | RE: st: Stata always breaks down when we use -nnmatch- |
Date | Fri, 26 Nov 2010 17:22:58 +0000 |
This question appears to assume perfect recall of a question from "Stata Chris" statachris@gmail.com http://www.stata.com/statalist/archive/2010-02/msg00660.html and of a reply from Austin Nichols austinnichols@gmail.com http://www.stata.com/statalist/archive/2010-02/msg00669.html These posts were several thousand posts ago on 15 February 2010! Any way, neither of those two persons has any responsibility for the program or for providing support on it. -nnmatch- counts as a user-written program from SJ and SSC. David Drukker (ddrukker@stata.com) is probably the best person to advise. Right now, he is probably enjoying Thanksgiving. -findit- reveals no public source of program code later than 2004, so whatever bit in February is still likely to bite, on the same computers. Nick n.j.cox@durham.ac.uk Tara McIndoe Calder =================== Hi Austin, Chris Has there been any further progress regarding solving this issue with nnmatch? I am able to run the estimator on a sub-sample (2,000 observations) of my dataset (17,000 observations) but above 2000 observations I run into memory issues, even on high memory computers. Any advice is most welcome. * * 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/