On Wed, 14 Sep 2011, Mauricio Cornejo wrote:
Thanks Z.
Let me clarify my problem a bit further ... as I don't think I could use
cut() or spine() to solve it.
I have a data frame with three columns (A, B, Value). 'A' and 'B' are
categorical and 'Value' is continuous and non-negative. I'm looking to make
a mosaic plot with A on one axis and B on the other. The size (i.e. area)
of the tiles would correspond to the proportion of the total Value
represented by the given combinations of the levels of A and B.
OK, that's easy. Just create the corresponding contingency table which has
non-integer cell entries and then call mosaic().
Consider this simple artificial example:
## artificial data
set.seed(1)
d <- data.frame(A = rep(1:2, each = 4), B = rep(1:2, 4), Value = runif(8))
d
## corresponding table
tab <- xtabs(Value ~ A + B, data = d)
## mosaic display
mosaic(tab)
hth,
Z
Thanks again,
Mauricio
____________________________________________________________________________
From: Achim Zeileis <[email protected]>
To: Mauricio Cornejo <[email protected]>
Cc: "[email protected]" <[email protected]>
Sent: Wednesday, September 14, 2011 1:45 PM
Subject: Re: [R] Can 'mosaic' be used with a continuous variable?
On Wed, 14 Sep 2011, Mauricio Cornejo wrote:
> Hello,
>
> I'm wondering if the 'mosaic' plot of the vcd package (or any other
function for that matter) can be used with a continuous variable that should
be represented via various categorical variables.? All the documentation
I've read lead me to believe that it only works with counts of categories.
>
> Alternatively, I've thought of first creating a contingency table where
the frequencies would really be the values of my continuous variable as
opposed to counts ... but I don't know how to do that.
Well, you can always use cut() to create a categorical variable from a
continuous variable. But you have to do that in advance before calling
mosaic().
For a bivariate display with a continuous x and a categorical y, you can use
spine() which does the categorization for you.
hth,
Z
> Any insight would be appreciated.
>
> Thanks,
> Mauricio
>
> [[alternative HTML version deleted]]
>
>
______________________________________________
[email protected] 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.
