Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
| From | Neil Shephard <nshephard@gmail.com> |
| To | statalist@hsphsun2.harvard.edu |
| Subject | Re: st: how to reshape data so that can perform calculations |
| Date | Tue, 22 Mar 2011 15:57:47 +0000 |
On Tue, Mar 22, 2011 at 3:10 PM, Kathleen Oberst <Kathleen.Oberst@hc.msu.edu> wrote: > I have a dataset with person ID, discharge date, and another calculated > field in a long format as follows: > > Id discharge date daysmissed y > x 1/1/2010 14 1 > x 2/1/2010 9 2 > y 1/5/2010 0 1 > z 2/8/10 25 1 > z 4/10/10 19 2 > z 9/30/10 8 3 > > > I'm trying to keep just those records of people who had >1 discharge and > then I want to calculate the elapsed time between dicharges. I generated > a marker 'y' to act as a counter for #admissions but can't really drop > anyone where y equals 1 because I will need that first d/c date for the > subsequent calculation. I've not had luck with reshape. You've not shown us how you created y, but one way to get it such that its the total number of observations for a given Id is.... bysort Id (dischargedate) : gen y = _N ...because _N is the total number of observations _within_ the defined group. You can then drop those individuals who only have one recorded discharge date with... drop if(y == 1) Useful reading would be... Cox NJ (2002) Speaking Stata: How to move step by: step. The Stata Journal 1:86-102 Neil -- “Truth in science can be defined as the working hypothesis best suited to open the way to the next better one.” - Konrad Lorenz Email - nshephard@gmail.com Website - http://kimura.no-ip.org/ Photos - http://www.flickr.com/photos/slackline/ * * 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/