R help - May 2008 - rlm and lmrob error messages

Hello all,

I'm using R2.7.0 (on Windows 2000) and I'm trying do run a robust 
regression on following model structure:

model = "Y ~ x1*x2 / (x3 + x4 + x5 +x6)"

where x1 and  x2 are both factors (either 1 or 0) and x3.....x6 are numeric.
The error code I get when running rlm(as.formula(model), data=daymean) is:

error in rlm.default(x, y, weights, method = method, wt.method = 
wt.method,  :
  'x' is singular: singular fits are not implemented in rlm

As singular.ok = TRUE is default with lmrob from the robustbase package, 
I tried running the model using lmrob, but I get following error message

Too many singular resamples
Aborting fast_s()

error in lmrob.S(x = x, y = y, control = control) :
  C function R_lmrob_S() exited prematurely

I don't understand what the error messages mean, especially in the case 
of lmrob and greatly appreciate any help or ideas as to what might cause 
the error. The model works fine with lm.

Kind regards,

Susanne

Susanne Komhard <susanne.komhard <at> ise.fraunhofer.de> writes:
> I'm using R2.7.0 (on Windows 2000) and I'm trying do run a robust 
> regression on following model structure:
> 
> model = "Y ~ x1*x2 / (x3 + x4 + x5 +x6)"
> 
> where x1 and  x2 are both factors (either 1 or 0) and x3.....x6 are
numeric.
> The error code I get when running rlm(as.formula(model), data=daymean) is:
> 
> error in rlm.default(x, y, weights, method = method, wt.method = 
> wt.method,  :
>   'x' is singular: singular fits are not implemented in rlm
Please supply a complete example, there is too much guessing involved here.
Assuming you use the function in MASS (there is one in limma), it could be that
the error comes from your "" around the model:


library(MASS)
summary(rlm("stack.loss ~ Water.Temp", stackloss))


Error in qr.default(x) : NA/NaN/Inf in foreign function call (arg 1)
In addition: Warning message:
In storage.mode(x) <- "double" : NAs introduced by coercion

summary(rlm(stack.loss ~ Water.Temp, stackloss))

Call: rlm(formula = stack.loss ~ Water.Temp, data = stackloss)
Residuals:
    Min      1Q  Median      3Q     Max 
-7.6816 -3.6368  0.3632  3.2736  8.7960 

Coefficients:
            Value    Std. Error t value 
(Intercept) -40.8262   8.5789    -4.7589
Water.Temp    2.7612   0.4024     6.8619

Residual standard error: 5.392 on 19 degrees of freedom
>

Dieter