Dear all
I hope I can explain what I am looking for ... I want to generate a two-dimensional histogram where
1. the widths of the bars vary (based on a "weighting") and
2. the length of each box within the bars displays frequencies of a categorical variable (similar across bars) AND
3. the color of each box within the bars has different color based on a color gradient

In essence (quote from help file of heatplot (SSC): "a plot of a trivariate distribution where the color gradient is used to visualize the value of z within bins of y and x".

#1 and #2 might be achieved by using spineplot from SSC
#3 can be achieved by using heatplot from SSC. Actually, heatplot allows customizing the binning of the y- and x-values but only with regular spacing.

The data I want to visualize is already provided in aggregated form. Here a made-up example (with 3 levels of x- and y-variable):
yvar yvar_size xvar xvar_size z (gradient)
A 70 1 10 0.53
A 70 2 30 0.89
A 70 3 7 0.21
B 30 1 10 0.12
B 30 2 30 0.65
B 30 3 7 0.25
C 10 1 10 0.95
C 10 2 30 0.47
C 10 3 7 0.69
... ... ... ... ...
Maybe I have overlooked something or did not understand the help files but I was not able to find anything (I also googled for ggplot2 in R and could not find anything). Maybe the graph does not make sense?

Any help would be appreciated.

Best
Sven