Hi: Use can use dcast() from the reshape2 package:
library(reshape2) > dcast(ft.df, Class + Sex + Age ~ Survived) Using Freq as value column: use value_var to override. Class Sex Age No Yes 1 1st Male Child 0 5 2 1st Male Adult 118 57 3 1st Female Child 0 1 4 1st Female Adult 4 140 5 2nd Male Child 0 11 6 2nd Male Adult 154 14 7 2nd Female Child 0 13 8 2nd Female Adult 13 80 9 3rd Male Child 35 13 10 3rd Male Adult 387 75 11 3rd Female Child 17 14 12 3rd Female Adult 89 76 13 Crew Male Child 0 0 14 Crew Male Adult 670 192 15 Crew Female Child 0 0 16 Crew Female Adult 3 20 Adding value = 'Freq' to the argument list of dcast() will get rid of the warning message on the console. HTH, Dennis On Thu, May 26, 2011 at 11:54 PM, Marius Hofert <m_hof...@web.de> wrote: > Dear Prof. Ripley, > > many thanks for your quick reply. > > A character matrix (although clearly not very elegant) would be no problem, > xtable deals > with that. > I tried as.data.frame() before, but if one wants to have the same rows > as in ft, one has to use additional commands (?): > ft # => 16 rows > as.data.frame(ft) # => 32 rows; different order > Is there a simple way to get the same order of the variables as in ft? > > Cheers, > > Marius > > On 2011-05-27, at 08:17 , Prof Brian Ripley wrote: > >> as.data.frame(ft) >> >> seems straightforward enough. >> >> I don't think you actually want a matrix, as they would have to be a >> character matrix and the ftable object is numeric. >> >> On Fri, 27 May 2011, Marius Hofert wrote: >> >>> Dear expeRts, >>> >>> What's the easiest way to convert an ftable object to a matrix such that the >>> row names of the ftable object are shown in the first couple of columns of >>> the >>> matrix? This is (typically) required, for example, when the final goal is >>> to print >>> the matrix via xtable. >>> >>> Below is a rather complicated example of how to do it... >>> >>> Cheers, >>> >>> Marius >>> >>> ## Goal: convert an ftable() to a (character) matrix including the row >>> names of >>> ## the ftable object as columns in the matrix (so that the matrix can >>> be >>> ## nicely printed with xtable() for example) >>> (ft <- ftable(Titanic, row.vars=1:3)) # ftable object >>> rn <- attr(ft, "row.vars") # pick out rownames >>> rn. <- rn[length(rn):1] # unfortunately, we have to (?) change the order >>> due to expand.grid() >>> g <- expand.grid(rn.) # build the 3 columns containing the row names >>> (g. <- g[,length(rn):1]) # change order back; now contains the same row >>> names as ft >>> (ft.mat <- as.matrix(ft)) # convert ftable object to a matrix >>> ## now, cbind g. and ft.mat >>> cbind(g., ft.mat) # => now the rownames are there twice! ... although >>> dim(ft.mat)==c(16, *2*) >>> ## class(g.) => okay, probably we meant: >>> (res <- cbind(as.matrix(g.), ft.mat)) >>> require(xtable) >>> xtable(res) >>> ______________________________________________ >>> 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. >>> >> >> -- >> Brian D. Ripley, rip...@stats.ox.ac.uk >> Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ >> University of Oxford, Tel: +44 1865 272861 (self) >> 1 South Parks Road, +44 1865 272866 (PA) >> Oxford OX1 3TG, UK Fax: +44 1865 272595 > > ______________________________________________ > 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.
