Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Maarten Buis <maartenlbuis@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: predict with xtreg |
Date | Thu, 8 Nov 2012 16:10:15 +0100 |
On Thu, Nov 8, 2012 at 3:53 PM, Carmen Blanco wrote: > According to your help the problem is because few variables in my > model has a missing value for the observations. And , in fact, I want > to say that my panel data is unbalanced. Do you know please how to > calculate the random error component (u) in my panel data with another > way? Why would you want to do so? Those observations with missing values weren't used, so -predict- is doing the right thing to ignore them. If you want to use the observations with missing values you'll have to make sure that -xtreg- can use those observations, e.g. -help mi-. Be warned: doing that right is a major project, so plan a couple of months minimum for that and ask yourself if 52 observations is worth a couple of months of your live... -- Maarten --------------------------------- Maarten L. Buis WZB Reichpietschufer 50 10785 Berlin Germany http://www.maartenbuis.nl --------------------------------- * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/faqs/resources/statalist-faq/ * http://www.ats.ucla.edu/stat/stata/