Question

How does one formulate a model for a nonlinear set of data?

Answer 1

I have to come up with two models for this set of data: (2,3) (4,5) (6,4) (8,6) (10,5) (12,12) (14,16) (16,22) (18,26) and (20,32).

If it were constant, I'd have no problem. I'm just utterly confused on how to go about this.

Answer 2

best fit line?

Answer 3

What exactly do you mean by 'best fit'?

Answer 4

It's what you're trying to do. Hard to believe you're doing a scatter plot without having ever heard of "best fit"

https://www.google.com/#output=search&sclient=psy-ab&q=line+of+best+fit&oq=line+of+best+fit&gs_l=hp.3..35i39j0l3.854.4859.0.5029.16.16.0.0.0.0.350.2534.0j15j0j1.16.0...0.0...1c.1.12.psy-ab.XtG3VxSouWY&pbx=1&bav=on.2,or.r_qf.&bvm=bv.46340616,d.dmQ&fp=6d02d4d3f553fbde&biw=1161&bih=905

OTOH, most lines of best fit appear to be linear, while yours appears to be elliptical/hyperbolic - which you acknowledged when you spoke of "a nonlinear set of data".

But no doubt, if I can estimate your scatter visually and run an elliptical section through it by eye, then there's certainly a method of fitting such a curve mathematically.