My experience is contrary to both your inclinations.
1. -reshape long- will make it easier to sort out date order, so you
need not worry about dates beforehand. (Besides when you say "merge",
you presumably don't mean -merge-, but something else.)
2. I see no loop being needed here. As before, I envisage -rename-
and -reshape-.
Nick
On Fri, Dec 7, 2012 at 7:36 PM, roland andersson <rolandersson@gmail.com> wrote:
> I will do that, but before changing to the long stucture I think it
> could be easier to merge the observation with the same date. This
> means that if to datecolumns are equal I would like to merge the
> corresponding content columns. How can I write a loop that compares
> every var9 var12; var12 and var15 and if equal merge var8 and var11
> and so on.
*
* 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/
