On Thu, Mar 1, 2012 at 12:07 PM, Doran, Harold <hdo...@air.org> wrote: > Typically this list doesn't support general statistical questions and > unfortunately I don't have a better recommendation. It may be more helpful > for you to work with a statistician than seek help here. > > My point is simply that quantile regression is not for modeling outcomes that > are quantiles. The fact that you have a dependent variables expressed as a > percentile doesn't mean that quantile regression is the appropriate approach. >
It's fairly widely true that percentiles or ranks behave probabilistically as if they were independent uniform random variables (though unfortunately it can be quite hard to prove in individual cases) so that might be where to start. -thomas -- Thomas Lumley Professor of Biostatistics University of Auckland ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
