topic How do I fit a 3-parameter hill model in Fit Curve? in Discussions

How do I fit a 3-parameter hill model in Fit Curve?

ceecee — Fri, 09 Jun 2023 00:59:45 GMT

Is there a way to fix the lower asymptote (parameter c) at 0 in the Logistic 4p Hill model in Fit Curve? Basically, I want the 3-parameter version of the Logistic hill model.

Re: How do I fit a 3-parameter hill model in Fit Curve?

MRB3855 — Wed, 29 Mar 2023 08:20:51 GMT

Doesn't look doable in the Fit Model platform. Definitely doable in the Nonlinear platform. In the model library of the nonlinear platform choose Logistic 4p. Then select "Make Formula" to create the formula in a column. Then go to that column and alter the parameters and/or formula be as needed to be the Log 4p Hill model. You can just remove "c" parameter from the model, or leave it in and set it to 0 when you run the non linear model.

Re: How do I fit a 3-parameter hill model in Fit Curve?

Mark_Bailey — Wed, 29 Mar 2023 15:02:05 GMT

Have you tried the Logistic 3P model in Fit Curve? I do not have a good example of data for such a model, but it looks like this:

Re: How do I fit a 3-parameter hill model in Fit Curve?

ceecee — Thu, 30 Mar 2023 11:46:10 GMT

Thanks! I didn't even know you could do that! Also this blog post helped https://community.jmp.com/t5/JMPer-Cable/Nonlinear-modeling-part-3/ba-p/42506