Share your ideas for the JMP Scripting Unsession at Discovery Summit by September 17th. We hope to see you there!
Choose Language Hide Translation Bar
Highlighted
Guy04
Level II

Generating model fit descriptive formula suitable for a scientific publication

HI 

 

I am using JMP® Trial 15.1.0  on a Mac.

I have been using Analyze>fit model. Personality: Generalised linear Model, Distribution: binomial, Link Function: probit. See image.

My "construct model effects" includes a nested factor.

 

How do I get the model description equation that I am using, such that I could publish in a scientific paper.

Does any body have any idea how to do this i would prefer and interactive rather than scripting solution please (if possible). Or even could tell me the equation.

 

 To be honest I am a bit lost and am only doing this analysis to help some colleagues out.

 

Thanks 

 

Guy

 

Using OS 10.14.6 

1 ACCEPTED SOLUTION

Accepted Solutions
Highlighted

Re: Generating model fit descriptive formula suitable for a scientific publication

Click the red triangle at the top of the platform and select the command that saves the prediction formula (fitted model). Then you can open the formula editor on the new column and see the expression. A screen shot of this form is probably not good enough quality for publication, but you can re-create it in a specialized equation editor, such as MathType.

Learn it once, use it forever!

View solution in original post

5 REPLIES 5
Highlighted

Re: Generating model fit descriptive formula suitable for a scientific publication

Click the red triangle at the top of the platform and select the command that saves the prediction formula (fitted model). Then you can open the formula editor on the new column and see the expression. A screen shot of this form is probably not good enough quality for publication, but you can re-create it in a specialized equation editor, such as MathType.

Learn it once, use it forever!

View solution in original post

Highlighted
Guy04
Level II

Re: Generating model fit descriptive formula suitable for a scientific publication

Thanks Mark that worked great.

I attach a picture of what I got when I followed your instructions.

Highlighted
julian
Community Manager Community Manager

Re: Generating model fit descriptive formula suitable for a scientific publication

Hi @Guy04

I like Mark's suggestion as it gives you a formula column and the text of the formula that you can then use in whatever software you choose to make more aesthetically pleasing. Another option is to use the same Red Triangle at the top of the report that Mark mentioned, then select Estimates > Show Prediction Expression. This will toggle on the prediction expression at the bottom of the report. Next, we'll want copy that prediction expression as a vector object. Right-click the grey triangle next to the Prediction Expression > Edit > Copy Picture. This will copy the outline box holding the prediction expression to your clipboard, and now you can paste anywhere you like. If you have Acrobat Pro, Illustrator, or another vector graphics software, you can edit the individual elements of this vector graphic and export and unlimited resolution.

I hope this helps!

@julian 

Highlighted
Guy04
Level II

Re: Generating model fit descriptive formula suitable for a scientific publication

HI Julian 

 

I tried to look for the "Estimates > Show Prediction Expression" menu but could not see it.  I attach what I could see.  Might  it be that this menu option is not available will all model options.

I attach the options that I did have .

Thanks for you help in anycase

Highlighted
julian
Community Manager Community Manager

Re: Generating model fit descriptive formula suitable for a scientific publication

Hi @Guy04

That's my mistake! I did not see that you were using the Generalized Linear Model personality within Fit Model. The option I suggested is not available in that personality, but will be for standard least squares models. 

Sorry about that!

@julian 

Article Labels

    There are no labels assigned to this post.