R help - May 2011 - nlrob(...) returns error message

Dear all,
?
I implemented a non-linear model using nls(...) and it works just fine. I now
tried to run the same model using nlrob(...) which basically does the same but
uses a more robust estimation procedure. My problem: I cannot seem to get
nlrob(...) running. Irrespective of how I try to call the function, I always get
the error message "Error in is.null(x) : 'x' is missing". I do
not use the variable "x" at all so this must be some kind of different
problem.
?
Can anyone help me out here?
?
Many thanks
?
Timo
?
Karlsruhe Institute of Technology
?
___________________________________________________________
Schon geh?rt? WEB.DE hat einen genialen Phishing-Filter in die

On May 16, 2011, at 4:19 PM, Timo Schlaefer wrote:
> Dear all,
>
> I implemented a non-linear model using nls(...) and it works just  
> fine. I now tried to run the same model using nlrob(...) which  
> basically does the same but uses a more robust estimation procedure.  
> My problem: I cannot seem to get nlrob(...) running. Irrespective of  
> how I try to call the function, I always get the error message  
> "Error in is.null(x) : 'x' is missing". I do not use the
variable
> "x" at all so this must be some kind of different problem.
>
> Can anyone help me out here?
Since 'x' is not an argument name for nlrob, it must be an argument  
for some other function call. Since you have not seen fit to include  
the code you used, the number of possible sources of error is  
denumerably infinite. You are the only one who can narrow that to a  
workable finite number. Perhaps with:

traceback()


-- 

David Winsemius, MD
Heritage Laboratories
West Hartford, CT