[R] Adding ranks to a repeatedly ragged array

[Prof Brian Ripley]

What do you mean by `a repeatedly ragged array': you haven't defined it?

There is no dataset `barley' in vanilla R.  Which one did you mean?
There is a data frame in package lattice, but that is neither an array nor 
ragged.

If that is what you want, rank.lists will be a list matrix.  The easiest
thing to do is to find out the corresponding row numbers and assign there.

pos <- with(barley, tapply(1:120, list(site=site, year=year), na.pass))
barley[unlist(pos), "ranks"] <- unlist(rank.lists)


On Mon, 13 Sep 2004, Wolfram Fischer wrote:

> How can I add an extra column containing the rank
> to a ragged array indexed by more than one grouping
> factors?
> 
> E.g. with the barley dataset: 
> How can I to add an additional column ``rank''
> containing the rank of the ``yield'' of
> the different varieties in relation to the indices
> ``year'' and ``site'' to the barley dataframe?
> 
> I achieved to calculate the ranks with:
> 	rank.lists <-
> 	with( barley, tapply( yield, list( site=site, year=year ), rank ) )
> but I do not manage to merge this result
> to the original dataframe ``barley''.

> PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html

Please do as that asks, including reading its references.  We should not 
have to guess at the meaning of your subject line, where you got a dataset 
from etc.

-- 
Brian D. Ripley,                  ripley at 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