Faq what is PLS1 v PLS2 and how to create separate PLS1 models from multi column y block

From Eigenvector Research Documentation Wiki
Revision as of 13:11, 5 December 2018 by imported>Lyle
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

Issue:

What is PLS1 vs PLS2 and how do I create separate PLS1 models when I have a multi-column y-block?

Possible Solutions:

If your y-block contains multiple y-columns, PLS always performs a PLS2. This creates a single model where all y-columns contribute to the loadings of the model. This can sometimes improve performance of the model, particularly when the signal in the different y-columns is correlated but the noise is not. A PLS1 model is a model built on a single y-column and the model reflects only covariance between the X block and that single y-column.

If you want separate PLS1 models for each y-block column, you have to exclude all but one of your y-columns and build independent models. To choose y-columns to use, see the Edit/Select Y-Columns menu item in the Analysis GUI or, from the command line, use the y-block's include{2} field. The resulting PLS model will be built for only the included columns.

This distinction is necessary because each PLS1 model contains a different set of loadings, Q, T2, and limits and an individual model can contain only one set of these.


Still having problems? Please contact our helpdesk at helpdesk@eigenvector.com