Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | rado645-bg@yahoo.de |
To | statalist@hsphsun2.harvard.edu |
Subject | AW: st: on construction of variables in Stata |
Date | Mon, 28 Mar 2011 17:14:18 +0100 (BST) |
Dear Nick, thanks a lot! This was very helpful! Best, Rado ----- Ursprüngliche Mail ---- Von: Nick Cox <njcoxstata@gmail.com> An: statalist@hsphsun2.harvard.edu Gesendet: Montag, den 28. März 2011, 9:17:46 Uhr Betreff: Re: st: on construction of variables in Stata If we knew these id-specific dates from a separate variable, this would be gen VAR = date > specific As it is, it seems you need clonevar VAR2 = VAR bysort id (date) : replace VAR2 = sum(VAR2) as missings are ignored by -sum()-, and this sum is bumped up by 1 when you hit 1. Another way to do it bysort id (date) : replace VAR2 = VAR2[_n-1] if missing(VAR2) replace VAR2 = 0 if missing(VAR2) That approach is documented FAQ . . . . . . . . . . . . . . . . . . . . . . . Replacing missing values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . N. J. Cox 2/03 How can I replace missing values with previous or following nonmissing values? http://www.stata.com/support/faqs/data/missing.html I do this on a clone in case I have misunderstood you. Nick On Mon, Mar 28, 2011 at 2:06 PM, <rado645-bg@yahoo.de> wrote: > I have a large data-set sorted by id and date over a period of three years (at > the daily frequency). > > > For each panel defined by id, I need to build a variable that takes a value of > one at a specific date that is different for each id. Consequently, the > variable (VAR) must take a value of 1 till end of the sample period and it >must > take a value of zero at any date before this specific day. > > > Currently, VAR has values of 1 only assigned to the specific days for each id > and missing observations elsewhere. > > > Could you please let me know if I could ask stata with some command to fill in > VAR with zeros preceding these specific days and to fill it in with ones > afterwards (ie after the occurrence of the specific date). * * 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/