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 | st: RE: Using lagged variables in panel data analysis |
| Date | Tue, 17 Jan 2012 16:11:34 +0000 |
It is usually better to create lagged variables with panel data using -tsset- or -xtset- followed by L. You will always have the problem that there is no value before the first. -drop-ping the first in each panel just makes your plight worse by throwing away some of your data. It can't solve this problem. On the whole, you need do nothing here: Stata's time series and panel commands do what they can with missing values, which usually means ignoring them. Nick n.j.cox@durham.ac.uk Edward James I have a question for conducting a panel data analysis. I make some lagged variables using "x[_n-1]" commands from 1980 to 2007 and want to conduct GLS. When I make lagged variables, is the case of the first period supposed to be "."? But it has a unknown value as you can see below. +----------------------------------------+ country year cpiannual lcpiann~l ---------------------------------------- 197. Germany 1980 5.441055 1.488074 198. Germany 1981 6.344242 5.441055 199. Germany 1982 5.241046 6.344242 200. Germany 1983 3.293413 5.241046 201. Germany 1984 2.405797 3.293413 ---------------------------------------- In this case, should I drop the case of 1980? * * 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/