Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: duration of STATA job


From   austin nichols <austinnichols@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: duration of STATA job
Date   Wed, 31 Aug 2005 12:01:31 -0400

I am guessing that your do-file has caused the process to hang.  If
you need to generate 40,000 dummy variables you will need about 2GB of
RAM and maybe 5 or 6 minutes, I think.  Perhaps the server is under
load from other processes run by other users, but not 18 hours' worth.
 How much other stuff is on your data?  Try an experiment like

set virtual off
set mem 3000m
use yourfile
g id=_n
qui compress
sort id
save yourfile2
keep id dummy1- dummy200 indicator1-indicator200
foreach d of varlist dummy1-dummy200 {
 foreach i of varlist indicator1-indicator200
  g byte d`d'x`i'=`d'*`i'
 }
}
sort id
save interactions
use yourfile2
merge id using interactions

On 8/31/05, Wilson, Nicholas <Nicholas_Wilson@brown.edu> wrote:
> I am running a program in STATA where I have a foreach loop inside of another foreach loop.  This generates all pairwise interactions of 2 sets of about 200 dummy variables each.  The number of observations is about 500,000.  I am running this remotely on a UNIX server with 16GB of RAM (I have requested 10GB for this job).  Do you have any idea how long this will run?  I can tell you that it has already run for about 18 hours and isn't done yet.  This doesn't bode well since this is just for year 1850 and I am running it for 1850, 1860, ..., 1930.

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   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   |   What's new   |   Site index