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; } } }

How do I save the eigenvalues from PCA into dataset?

gemmahodgson

I am using JMP Pro 13 doing PCA and although I can see all the output I need on the screen, I want to be able to extrct the eigenvalues and use them as weights for another anlaysis, so I need them in a dataset, but there doens't appear to be any option to get them out. The only thing I can save is the principal components...does anyone know how you can save the other parts of the output? Either via scripting or interactive is fine.

Thanks in advance

Gemma

Accepted answers

All comments

Dan_Obermiller

Try right-clicking on the table and choose Make into Data Table.

gemmahodgson

Thanks for your quick answer Dan, I found that, however I need to do it lots of times (as I have to do 12 PCAs) so I wanted to run it in JSL as a script, where the PCA is done (by consumer) and the eigenvalues exported each time...any ideas on that?

txnelson

Here is a simple example of a script creating the Eigenvalue data table

names default to here(1);
dt=open("$SAMPLE_DATA/semiconductor capability.jmp");

pc=dt << Principal Components(
	Y( :NPN1, :PNP1, :PNP2, :NPN2, :PNP3 ),
	Estimation Method( "Default" ),
	"on Correlations",
	Eigenvalues( 1 ),
	SendToReport(
		Dispatch(
			{"Summary Plots"},
			"PCA Summary Plots",
			FrameBox,
			{Frame Size( 43, 40 )}
		),
		Dispatch(
			{"Summary Plots"},
			"PCA Summary Plots",
			FrameBox( 2 ),
			{Frame Size( 53, 40 )}
		)
	)
);
report(pc)["Eigenvalues"][1][1]<<make into data table;

jules

Hi Gemma,

If I'm understanding your question correctly you want to save the table of eigenvalues shown in the JMP PCA report for all the extracted components. If that's right, a simple way to do this is to right-click the table in the report and select "Make into Data Table."

I hope this helps!

Julian

jules

Oops, I see now that Jim already responded with a great method for handling this across many reports!

gemmahodgson

That's great thank you so much - I will try it now! Can't believe all the v quick replies - I was expecting to wait weeks!

louv