piecewise fitting with user defined functions

I have a data set defined by vectors [x] and [y] to which I want to perform a fit using a function that I define. This function (call it f(x)) has three open parameters (a, b, c). The left half of the data will have one set of parameters (a1, b1, c1), and the right half of the data set will have a different set of parameters (a2, b2, c2). Is there a way to perform such a piecewise fit and obtain a single model (call it g) that I can use similar to the output of:

cfun = fit(xdata,ydata,libname)?

Answer 1

[per isakson]

See

• SLM - Shape Language Modeling by John D'Errico
• BrokenStickRegression by Peter Nave

In the File Exchange

/ per