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: lag vars |
Date | Tue, 15 Feb 2011 13:39:11 +0000 |
Not necessarily. bysort pid (time) : gen poor0 = poor[1] More importantly, there are all sorts of logic tangles opening up here given that -poor0- and -poor- have inbuilt overlap. Nick n.j.cox@durham.ac.uk Sabrina Carrossa I found the solution: - poverty status in time 0--> by pid: gen poor0 = poor[1] // poverty status at start Thanks! On 15 February 2011 13:23, Sabrina Carrossa <carrossa@gmail.com> wrote: > I have a three waves panel Db: > 2004 (t0) > 2005 (t1) > 2006 (t2). > > I wish to use a probit model (xtprobit) to analyze the TSD of poverty, > and I need to specify some lagged variables such as: > - the lagged poverty status > - the poverty status in time 0 > - the current employment status (employed or not) > - the employment status in t2 > > I know that I can refer to the lagged poverty status by typing: > "l.poor", but what about the other variables? > Should I compute them by gen/egen commands? Is there a more handy and > quick way to obtain them? * * 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/