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: byable returns : return list after by prefix
From
László Sándor <[email protected]>
To
[email protected]
Subject
Re: st: byable returns : return list after by prefix
Date
Sat, 24 Aug 2013 14:33:56 -0400
I loved Austin's idea for writing my own wrapper for -sum, meanonly-
with stored means, as -bys: sum, meanonly- is still by far the fastest
to calculate what I need, though it is so silent I could never
cross-check the values. In any case, in my test this (-mymns- from
above) does not run in the -bys loop- because the op.sys. refuses to
give me memory. I am not sure how that can be, I requested 240 GB and
the data takes "only" 20 GB…
Any thoughts are welcome.
On Fri, Aug 23, 2013 at 12:58 PM, László Sándor <[email protected]> wrote:
> It is disappointing and a bit surprising that statsby ran at least
> five times slower than the -bys- itself with the same command.
>
> But there we go.
>
> -tabstat- is even slower. I am not testing Mata at the moment. It
> would be a bit surprising if that simple code could be better
> parallelized than all the -by- machinery. But maybe…
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/