Re: Constrained non linear regression using ML

[Dupuis]

Dear Octave users,

I have to fit the non linear regression:

y~1-exp(-(k0+k1*p1+k2*p2+ .... +kn*pn))

where ki>=0 for each i in [1 .... n] and pi are on R+.

I am using, at the moment, nls, but I would rather use a Maximum
Likelhood based algorithm. The error is not necessarily normally
distributed.

Looks like a Prony problem, isn't it ? Are the pi regullary spaced or not ?
In case of equal spacing, I devised a method based upon Prony method. The
algorithms are written in Octave, but I'm missing time to distribute them.
Maybe you could try them ?

Regards

Pascal
-- 
View this message in context: 
http://old.nabble.com/Constrained-non-linear-regression-using-ML-tp27923049p27930870.html
Sent from the Octave - General mailing list archive at Nabble.com.