[RsR] robustbase::lmrob() difficulty continued

[Ajay Shah]

Folks,

A few days ago I had asked a question
https://stat.ethz.ch/pipermail/r-sig-robust/2008/000208.html and got a
tonne of help. One issue that was pointed out was recoding a factor so
as to not have a missing level in it. Another issue that was pointed
out was that excluding an intercept seems to tickle a bug.

A few weeks later, various packages have revved. I have solved the
problem of factor coding. Now the intercept exclusion no longer seems
to be an issue. Now MASS:rlm() works for me as does
robust::lmRob().

But robustbase::lmrob() still does not work (robustbase 0.4-3). Here's
a demo:

---------------------------------------------------------------------------
> load(url("http://www.mayin.org/ajayshah/tmp/demo.rda"))
> library(robustbase)
> lmrob(spec.full, data=tmp)

Too many singular resamples
Aborting fast_s_w_mem()

Error in lmrob.S(x = x, y = y, control = control) : 
  C function R_lmrob_S() exited prematurely
Calls: lmrob -> lmrob.fit.MM -> lmrob.S
---------------------------------------------------------------------------

I'm keen to use robustbase::lmrob() in order to get the right
inference (see
https://stat.ethz.ch/pipermail/r-sig-robust/2008/000217.html). Do you
have a sense of how this might be achieved?

-- 
Ajay Shah                                      http://www.mayin.org/ajayshah  
ajayshah using mayin.org                             http://ajayshahblog.blogspot.com
<*(:-? - wizard who doesn't know the answer.