Peter, thanks for your response! The problem is not how indexing works, but rather the question what is being indexed here. If I understand the description correctly then it is wrong. In the special and common case where all possible levels do actually occur in the data frame it coincidentally happens to work the way it is described but not when the data frame contains only data points for some of the levels. Then it appears that the indexing vector has to be bound to 1:length(unique(f)) which is unequal 1:nlevels(f). (f is a factor here.)
Titus On Wed, Nov 25, 2009 at 10:32 PM, Peter Ehlers <ehl...@ucalgary.ca> wrote: > > Titus Malsburg wrote: >> >> The documentation of xyplot could be improved here. It says: >> >> "If 'index.cond' is a list, it has to be as long as the number of >> conditioning >> variables, and the 'i'-th component has to be a valid indexing vector >> for the >> integer vector '1:nlevels(g_i)' (which can, among other things, repeat >> some >> of the levels or drop some altogether)." >> >> It should make explicit that nlevels is the number of levels actually >> used in the data and not length(levels(f)). >> > It does say "... _valid_ indexing vector ..." (my emphasis). > If nlevels(g) = 5, but you're only plotting 3 panels, it seems > to me that c(3,1,5) might be a valid indexing vector. > > -Peter Ehlers > >> Cheers, >> >> Titus >> >> ______________________________________________ >> 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. >> >> > ______________________________________________ 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.
