cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Try the Materials Informatics Toolkit, which is designed to easily handle SMILES data. This and other helpful add-ins are available in the JMP® Marketplace
Choose Language Hide Translation Bar
bwanders
Level III

Exponential Curve Fitting

Look at my attached file.  I have fitted an Exponential 3P curve to data that is based on %.  I have added a range check as well as a response limit for my "% knockdown" column.  I would like to have my fit curve parameter estimates not exceed 100%.  Is there a way to do this?

1 ACCEPTED SOLUTION

Accepted Solutions
mpb
mpb
Level VII

Re: Exponential Curve Fitting

Instead of using the Fit Curve personality of Nonlinear, you can use the Custom Model personality (Help Index: Nonlinear Platform, Custom Models).

1. Create a column formula (say in column MyFormula) with parameters a, b and c:  a + b*exp(c*Time)

2. In the initial Nonlinear dialog Put MyFormula in the X, Predictor Formula box and %Knock Down in the Y, Response box. Then click OK.

3. In the second Nonlinear dialog you can click the red triangle by Nonlinear Fit and choose the first item, Parameter Bounds. You will then have the opportunity to supply Lower and/or Upper bounds for each of a, b, and c.

View solution in original post

2 REPLIES 2
mpb
mpb
Level VII

Re: Exponential Curve Fitting

Instead of using the Fit Curve personality of Nonlinear, you can use the Custom Model personality (Help Index: Nonlinear Platform, Custom Models).

1. Create a column formula (say in column MyFormula) with parameters a, b and c:  a + b*exp(c*Time)

2. In the initial Nonlinear dialog Put MyFormula in the X, Predictor Formula box and %Knock Down in the Y, Response box. Then click OK.

3. In the second Nonlinear dialog you can click the red triangle by Nonlinear Fit and choose the first item, Parameter Bounds. You will then have the opportunity to supply Lower and/or Upper bounds for each of a, b, and c.

bwanders
Level III

Re: Exponential Curve Fitting

Thank you.  This nonlinear custom model definitley brought me to a new place in JMP.