Bi-linear Regression- Model fitting

Hi every body I simply have a response variable (y) that I am trying to predict with one explanatory variable (x) but the shape of the scatter plot is distinctly bi-linear. It would be best described by two straight lines. Is there a way of fitting a linear model to give me a bi-linear fit? Is there any command in Matlab for this kind of regression?

Answer 1

[Image Analyst]

See attached demo, that I've posted many, many times before. The algorithm is I divide the data into two sections (left and right) and fit lines to each section. Then I find the point where the slope difference is greatest. This is where the data should be split.

Answer 2

[Ahmet Cecen]

That sounds like a problem that would be better solved with a clustering or classification tool rather than regression. If you really need the individual lines, I would first do a classification run followed by linear regression.

Answer 3

[Ameer Hamza]

bisection method