robustbase problem [bug?] in adjbox function.

hello list,

i'm a bit puzzled by the error message i get when i copy past this in R:
data<-c(16,13,17,4,15,24,59,18,33,8,42,19,20,4,11,9,3,7,10,3,3,67,4,4,13,16,6,3,3,6,3,4,35,10,16,11,24,7,47,8,8,2,12,3,8,4,3,6,6,10,2,9,3,15,21,13,8,16,2,5,14,9,21,4,9,11,36,4,8,4,6,4,10,13,11,5,4,16,14,3,22,20,5,2,8,12,14,27,2,13,8,18,18,6,31,3,14,23,13,27,13,6,7,10,12,3,17,30,3,13,13,9,29,4,4,18,7,23,4,20,3,9,4,4,3,3,5,32,11,3,4,5,10,12,7,4,13,10,11,37,6,8,6,3,12,8,10,6,12,7,8,9,5,6,12,5,4,24,7,2,9,2,4,3,8,5,10,5,4,4,7,4,7,7,4,16,4,3,12,4,3,10,27,4,14,7,10,4,9,8,3,5,3,8,5,2,8,38,3,3,3,4);
plot(x=c(1,2),y=rep(1,2),ylim=range(data),xlim=c(0.5,2.5),type="n",main=c("Adjusted
boxplot vs. Classical boxplot"),ylab=c("Loss Data"),xlab=c(""),axes=F)
adjbox(x=datac,pars=list(outpch=16,outcex=2,outcol="light
blue"),add=T,at=1,col="light blue")
boxplot(x=datac,pars=list(outpch=16,outcex=2,outcol="light
green"),add=T,at=2,col="light green")
axis(2,at=seq(0,max(data)+10,by=10),col="orange",labels=F)
axis(1,at=c(1,2),col="orange",labels=c("Adjusted boxplot","Classical
boxplot"))

in my computer adjbox says:

"maximal number of iterations (100 =? 100) reached prematurely
Error in mc.default(x, na.rm = TRUE) :
   mc(): not 'converged' in 100 iterations"

R2.12 and the latest version of robustbase -in ubuntu 10.10).

On 2010-12-07 12:01, kv wrote:

hello list,

i'm a bit puzzled by the error message i get when i copy past this in R:

in my computer adjbox says:

"maximal number of iterations (100 =? 100) reached prematurely
Error in mc.default(x, na.rm = TRUE) :
? mc(): not 'converged' in 100 iterations"

You're not exactly providing a 'minimal' example. (We don't need
to know your colour preferences and plot titles, etc, since
those have no relation to the problem.)

The problem is with mc() which adjbox() calls (as is stated
on the help page for adjbox).

I suspect that there may indeed be a bug in the C code called
by mc(), possibly related to ties. I saw no comments about
ties on the help page for mc().

A workaround for you might be to add a very small disturbance
to your data, e.g. rnorm(length(datac), sd = 1e-10).

Note to the maintainer:
Here's a slightly simpler reproducible example:

set.seed(15)
x <- sort(sample(67, 101, TRUE))
mc(x)
# maximal number of iterations (100 =? 100) reached prematurely
# Error in mc.default(x) : mc(): not 'converged' in 100 iterations

Eliminating any one of the elements of x with index in
c(17:18, 33:45, 48:50) results in convergence.

Peter Ehlers

R2.12 and the latest version of robustbase -in ubuntu 10.10).

Thanks Peters, what i suspected (i.e. ties). For what it's worth i would add
that the problem happends when length(data)>100 (i.e. the second condition
in the mc.default() function).

Best,

On 2010-12-08 06:39, kv wrote:

Thanks Peters, what i suspected (i.e. ties). For what it's worth i would add
that the problem happends when length(data)>100 (i.e. the second condition
in the mc.default() function).

Best,

I don't think that the length of 100 is special (other
than in the 'doReflect' default).

Here are a couple of short vectors that cause the
error message:

   x1<- c(13, 30, 30, 32, 39, 40, 40, 40, 42)
   x2<- c(11, 31, 39, 39, 42, 46, 49, 49, 50)

   mc(x1, doReflect=FALSE)
   mc(x1, doReflect=TRUE)
   mc(x2, doReflect=FALSE)
   mc(x2, doReflect=TRUE)


Peter Ehlers