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; } } }

775037_en-US

liliconcom

set alpha level in Fit Curve

I had a customer ask if they could set the alpha level to get different confidence levels on the confidence intervals produced by the profiler for Fit Curve. I can't see anywhere that I can do that. I have found it for Fit Model and for Nonlinear, but not Fit Curve.

In the Fit Curve platform, you have the option of saving a formula column to the data table that Nonlinear can use. That is a possible work around but in this case, the customers data set is not producing a standard error from Nonlinear and so it can't get confidence intervals. The profiler in Fit Curve is able to produce 95% confidence bounds for the data, but Nonlinear can't produce anything.

Fit curve can save a formula column for the predicted response and a formula for the standard error. Then, the user can create columns with the confidence bounds desired. That is a work around but requires more effort from the user. I think there should be a way to do that more easily.

Any suggestions?

Re: set alpha level in Fit Curve

Thanks Charlie, this is not currently available but is a reasonable request. Captured in this Jira: https://jmpdiscovery.atlassian.net/browse/JMP-22313.