[R] Quantiles of data in a contingency table

[Matt Mohebbi]

On Thu, 17 Mar 2005 10:29:16 +0200, Markus Jäntti <markus.jantti at iki.fi> wrote:
> Matt Mohebbi wrote:
> > Hello,
> >
> > I have data of the following form:
> >
> >
> >>data <- data.frame(type=c("c","d","e"), size=c(10,20,30), count=c(20,10,5))
> >>data
> >
> >   type size count
> > 1    c   10    20
> > 2    d   20    10
> > 3    e   30     5
> >
> > I would like to compute the quantiles of size given the counts. For
> > instance, in this example, the median size would be 10. Is there an
> > easy way of doing this?
> 
> One at least is to the function wtd.median [and wtd.quantile] in package
> Hmisc by Frank Harrell.
> 
> install.packages("Hmisc")
> library(Hmisc)
> wtd.median(data$size, data$weights)
> 
> is likely a route to get you what you want.
Thanks. This worked great. 

I now would like to do a boxplot on a subset of this data. One way of
doing this would be to create a repeated entry form of the above data
frame. This would look like:
   type size 
 1    c   10  
 2    c   10  
 3    c   10  
 4    c   10   
 5    c   10    
 6    c   10    
 7    c   10    
 8    c   10    
 9    c   10    
 10    c   10    
 11    c   10  
 12    c   10  
 13    c   10  
 14    c   10   
 15    c   10    
 16    c   10    
 17    c   10    
 18    c   10    
 19    c   10    
 20    c   10    
(and similarly for types d and e)

I cannot seem to find this in R or Hmisc. Any ideas? 

Thanks,
Matt



> regards,
> 
> markus
> >
> > Is there a good way to deal with data in this format in general? Much
> > of R seems to center around having an entry for each item. This
> > question (http://www.r-project.org/nocvs/mail/r-help/2000/0102.html)
> > seems to be related but no one provided an answer.
> >
> > Thanks,
> > Matt
> >
> > ______________________________________________
> > R-help at stat.math.ethz.ch mailing list
> > https://stat.ethz.ch/mailman/listinfo/r-help
> > PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
> >
> 
> --
> Markus Jantti
> Abo Akademi University
> markus.jantti at iki.fi
> http://www.iki.fi/~mjantti
>