[Help-gsl] Non-linear Multi-parameter Least-squares Fitting

[Gordan Bobic]

Hi,

I apologize for asking a question that is arguably documented in the
manual, but I'm finding the manual quite difficult to follow on the
subject in question with the notation used. So, I'm hoping somebody
could help me out with an example.

I've got some sampled data I need to analyze for oscillations (sampled
machinery vibration). I'm trying to fit an equation to it, of the
following form:

Y = a * sin (bX + c) + d

a = amplitude
b = frequency
c = offset along the X axis
d = offset along the Y axis

I wrote an algorithm that guesstimates a reasonable starting point for
the gradient descent, but I'm a bit lost as to how I can get GSL to do
this for me.

Can anyone help me out with a short example to help me get to the
learning curve?

Thanks.

Gordan