I am coming from my earlier question on linear fit solved here.
In my actual case all the data points do not strictly fall in straight line (which is expected from theory). Please see attached where the smoothing blue line is a guide to the eye.
I would like to fit a straight line to the upper 4 (or 5) points and separately a straight-line to the 3 (or 2) points and get the equation (or slope) for each. How to best implement this (perhaps with both lines visible)?
When it's too good to be true, it's neither