[R] Same regression per sub-group: apply?
ggrothendieck at gmail.com
Fri Dec 7 13:41:32 CET 2007
On Dec 7, 2007 3:32 AM, Rense Nieuwenhuis <r.nieuwenhuis at student.ru.nl> wrote:
> Dear helpers,
> I've come up with what is probably a simple problem, but I cannot
> find the solution. I have a data-set containing survey-data from
> several countries. What I want to do is to perform some regression
> analyses, for each country separately. The question is, how to do
> this nicely (thus without repeating the same syntax with another
> `subset' argument).
> I thought of the following:
> model.per.country <- tapply(data, data$COUNTRY, function(x) glm
> (dependent.var ~ FEMALE + AGE + EDUCLIN + (), family=binomial,
> But this does not work. What goes wrong, I think, is that the
> dependent variable is clustered according to `Country', but not so
> for the predictors. The error message I received:
> Error in tapply(dat, dat$COUNTRY, function(x) glm(participate ~
> FEMALE + :
> arguments must have same length
> Could you please help me solving this little problem?
> Thanks in advance:
> Rense Nieuwenhuis
> [[alternative HTML version deleted]]
> R-help at r-project.org mailing list
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
More information about the R-help