[R] bug in nls?

[Berwin A Turlach]

G'day Petr,

On Thu, 26 Jun 2008 13:57:39 +0200
Petr PIKAL <petr.pikal at precheza.cz> wrote:

> I just encountered a strange problem with nls formula. I tried to use
> nls in cycle but I was not successful. I traced the problem to some
> parse command.
> 
> [...]
> 
> 
> I am not sure if this behaviour is a bug or feature.  [...]

It is definitely a feature.  

It is an error to believe that all modelling functions that use
modelling formulae use the same syntax for their modelling formulae.
This is, perhaps, easiest realised by observing how lm() and nls()
interpret "*" and "/" in model formulae.

In nls(), "[..]" can be used to index parameters, if the parameter is
allowed to change between groups in the data.  This seems to be a
little known feature, though there is an example that uses that feature
in MASS.  The contributed documentation "An Introduction to R: Software
for Statistical Modelling & Computing" by Petra Kuhnert and Bill
Venables, available from CRAN, also has such an example on pages 134
 and 230.  

The fact that nls() allows you to use "[..]" to index parameters in the
model formulae seems to conflict with the way you wanted to specify
the observed values in the formula.  I guess Gabor's solution is a fix
for your problem.

Cheers,

	Berwin

=========================== Full address =============================
Berwin A Turlach                            Tel.: +65 6515 4416 (secr)
Dept of Statistics and Applied Probability        +65 6515 6650 (self)
Faculty of Science                          FAX : +65 6872 3919       
National University of Singapore
6 Science Drive 2, Blk S16, Level 7          e-mail: statba at nus.edu.sg
Singapore 117546                    http://www.stat.nus.edu.sg/~statba