701488_en-US

liliconcom

PLS Interpretation

Hello,

 

I'm not very familiar with the partial least squares model and so I'm having trouble interpreting the results in jmp. For  example, if I open the "Prostate cancer" table available in JMP's sample index, and run the attached script "Fit Partial Least Squares":  How can I interpret these coefficients below? Are they the coefficients of the "factors" variables created in the factor regressions? 

 

 

Advanced Statistical ModelingPredictive Modeling and Machine LearningWindowsRe: PLS Interpretation

Before jumping to that particular report, I recommend spending some time getting to know PLS in general.  There is a lot of information available in those models, but it takes a little time to understand it.  Coefficients are rarely in the first few places I look.   Here is a paper that seems to walk through the platform with an analysis that might help:

https://support.sas.com/content/dam/SAS/support/en/books/discovering-partial-least-squares-with-jmp/65346_excerpt.pdf

Re: PLS Interpretation

To add just a bit to @ih 's advice...here's a link to an on demand webinar I hosted many years ago during my tenure as a JMP Senior Systems Engineer. The JMP Pro version is somewhat dated nowadays but the content hopefully is still relevant: Partial Least Squares: What to Use When Ordinary Least Squares Just Won't Work 

Re: PLS Interpretation

I can't mark my reply as a solution. Only the person who initiated the Discussion entry can mark it as a solution.

Re: PLS Interpretation

Or a super user...marked as solved.  Thanks Pete.

Re: PLS Interpretation

Thank you! I just presumed solution could only be granted by the original author.

Re: PLS Interpretation

Thank you sir for answering, you made my day. I appreciate you. Please mark it as a solution.

View more...