optim: curvefit_stat and fixed parameters

[Markus Appel]

Hello,

my goal is to calculate covp and corp for an optimization with 
nonlin_curvefit using curvefit_stat. After having some problems I tried 
to put together a minimal working example to reproduce the error (which 
may be a bug). I am not quite there yet, but encountered another problem:

The attached script uses nonlin_curvefit and curvefit_stat, one time 
with a parameter vector and another time with a parameter struct. I fit 
a polynomial with 3 parameters to some data, but added a 4th one on 
which the model function does not depend (for demonstration). Declare 
the 4th parameter fixed, and curvefit_stat (for parameter vector) gives:

> error: __dfdp__: product: nonconformant arguments (op1 is 4x1, op2 is 3x1)
> error: called from:
> error:   /home/mappel/octave/optim-1.3.0/private/__dfdp__.m at line 88, column 7
> error:    at line -1, column -1
> error:    at line -1, column -1
> error:   /home/mappel/octave/optim-1.3.0/private/__residmin_stat__.m at line 
> 548, column 14
> error:   /home/mappel/octave/optim-1.3.0/curvefit_stat.m at line 61, column 7
> error:   /home/mappel/test/fit_test.m at line 19, column 6

or correspondingly (for parameter structure):

> error: cell2struct: number of FIELDS does not match dimension
> error: evaluating argument list element number 1
> error: called from:
> error:    at line -1, column -1
> error:    at line -1, column -1
> error:   /home/mappel/octave/optim-1.3.0/private/__dfdp__.m at line 26, column 7
> error:    at line -1, column -1
> error:    at line -1, column -1
> error:   /home/mappel/octave/optim-1.3.0/private/__residmin_stat__.m at line 
> 548, column 14
> error:   /home/mappel/octave/optim-1.3.0/curvefit_stat.m at line 61, column 7
> error:   /home/mappel/test/fit_test.m at line 31, column 7

Don't keep the 4th parameter fixed, everything works fine and covp/corp 
contains NA for the 4th parameter which is reasonable. Try to declare 
parameter 3 as fixed (on which the model function actually depends and 
dfdp is not zero) and it works as expected, without error and more NA's 
in covp/corp.

Is this a bug or am I not using these functions as intended?

Thanks,
Markus

PS: As the error logs show, I use optim 1.3.0 (just updated all 
packages), this is on Ubuntu 12.04 with octave 4.1.0+ (sorry, forgot 
changeset).

fit_test.m
Description: Text Data