Choose Language Hide Translation Bar
Highlighted
RBAero
Level II

Missing P Values in Effect Summary OF Fit Model

Hi, I am using JMP-14.

I was using Fit Model and somehow my Effect Summary Dialogue is missing Data, no P-Values for the data.

Is there a reason why this would happen?

Thanks in advance 

 

RBAero_0-1591391345311.png

 

7 REPLIES 7
Highlighted
Thierry_S
Level VI

Re: Missing P Values in Effect Summary OF Fit Model

Hi, it is difficult to figure out what may have gone wrong without more details but the first thing that comes to mind is the large number of effect in your model. I wonder if your data coverage may not be sufficient to explore all these factors.

Would it be possible to share an example of the data you are working on? Also, are there key factors that would be worth exploring first? Finally, have you considered the Screening platform to priotitize the factors in your model.

Best,

TS
Thierry R. Sornasse
Highlighted
statman
Level VII

Re: Missing P Values in Effect Summary OF Fit Model

If your model is saturated (all DF's are accounted for in the model), there are no DF's to estimate the error term and therefore there are no tests of significance.

Highlighted
huangshans
Level II

Re: Missing P Values in Effect Summary OF Fit Model

Hi I have a similar problem and I don't think my model is saturated because there are 28 experiment and 6 variables. The two missing p values correspond to the two categorical variables. Why?

 

JMP effect summary.png

Highlighted

Re: Missing P Values in Effect Summary OF Fit Model

What does the Analysis of Variance, Parameter Estimates, and Effect Tests tables show?

Learn it once, use it forever!
Highlighted
huangshans
Level II

Re: Missing P Values in Effect Summary OF Fit Model

JMP response.png

Highlighted

Re: Missing P Values in Effect Summary OF Fit Model

So this indicates that you do not have adequate data to estimate your model. Fitting a model requires that you have enough data, but also you need to have the data collected in the proper locations. Think of it this way: I want to estimate a straight line relationship between Y and X. I collect 100 data points, but X is always 10. Lots of data, but I cannot estimate the line.

 

In your situation, I believe you also have a Singularity Details report. If you open that up you will see that you have a relationship (or confounding) between two of your input variables. It looks like it is between Screen Plate and SDS. JMP cannot estimate both of those effects because there is not sufficient data to do so. Make a plot of these two input variables. What does it look like? I think that might show you the issue.

Dan Obermiller
Highlighted
huangshans
Level II

Re: Missing P Values in Effect Summary OF Fit Model

Thanks, Dan. This is very helpful.