nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

What is "By ID" Column Property?

Martin

This is what I am doing:

1. I run a "Fit Curve" analysis of non-linear data using both a Grouping variable and a By variable (there are 5 of these)

2. Then apply a "Fit Logistic 4P" to the Fit Curve analysis

3. Then "Save Parameter Prediction Formula"

--> this step saves two items to a new column in my data table, (1) a Formula that I fully understand and (2) a "By ID" column property that I don't understand.

Does anyone know that this "By ID" column propery is and what does it represent? If seems to have a single large integer in it.

Thank you

Accepted answers

Mark_Bailey

The documentation that you cited shows an example in which the Group role is used. That role is different than the By role. The prediction formula that you saved is using the By role, which Nonlinear does not recognize. You need a parametric model to use with Nonlinear.

All comments

Mark_Bailey

You used Fit Curve with a variable in the By analysis role. The formula needs help to know which estimates are used with with By group (level). That information is stored in the column properties that you discovered.

It's documented.

Martin

My By variable is like "fruit", apple, orange, pear, etc. How does JMP translate that information into the integer value shown? And, where is it documented, I would like to understand this.

The reason why, is I do not believe JMP is behaving correctly with this information.

Martin

What is happening, is that each of my By levels (fruit) does show up in the formula "correctly". But, when I take that column and use it in Fit -> Nonlinear, all of the Parameters for each By level are showing up exactly the same, as though they all came from only one of the By levels. They are all different for each By level in the Formula that was saved from the Fit platform.

Mark_Bailey

First all, I apologize. I cannot find the documentation for this By ID property. I can find the documentation for the Predicting property:

Predicting

JMP automatically assigns the Predicting column property when you fit a model to a continuous response and save the prediction formula or prediction values. The Predicting column property identifies the platform used to create the prediction formula. That platform is listed as the Creator in the Model Comparison platform.

I do not know how the other property is used or what its value means.

The prediction formula is correct and functions properly. Why do you think that JMP is not working?

The prediction formula with a by variable is for prediction. It might not be intended to serve as a custom non-linear model to be fit in Nonlinear. I cannot launch Nonlinear with this prediction formula in the model role.