<>
Reto said
I have a dataset with about 100 dummy variables for companies (say
c1-c100) and another 100 dummy variables for Month-year-combinations
(say mo1-mo100). Now I would like to create 10'000 dummy variables
labeled c1mo1-c100mo100 and their values c1*mo1 to c100*mo100.
Unless you have Stata/SE or Stata/MP with plenty of RAM on your
machine, you cannot do this at all (see -help limits-). But why would
you want to do it anyway? Normally if you want to deal with two-way
fixed effects (by company and time period) you use a set of company
dummies and a set of time dummies (usually by allowing xtreg,fe to
generate the former automatically). But if you create a dummy that
identifies each (i,t), you can't very well include the set of those
dummies in a model, because there will be one for every observation.
Kit Baum, Boston College Economics and DIW Berlin
http://ideas.repec.org/e/pba1.html
An Introduction to Modern Econometrics Using Stata:
http://www.stata-press.com/books/imeus.html
*
* 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/
