PLS, OPLS, O2PLS

[FrancescaZ]

Hello everyone.
I am new to PLS techniques, and I am trying to study PLS applied to dimensionality reduction. I need to understand the basic PLS algorithm and the subsequent developed OPLS and O2PLS. Although I have read many articles, I still find it hard to understand the difference bewtween PLS and PLS-regression.

Can anyone explain this to me? Does anyone have any suggestion regarding which algorithm should be applied for dimensionality reduction?

Thanks in advance
F.

[Hengkov]

Hello Francesca,
PLS-PM for test complex model with multi-construct and multiple indicators.
PLS-Regression for handle multicollinearity with optimal predict.
Next, for test dimensionality used PCA and check item reliability.

Regards,
Hengky

[FrancescaZ]

Thank you Hengky.

I have now studied the NIPALS algorithm for PCA and PLS. I know that the same algorithm is used for OPLS and O2PLS.

Have you heard of these techniques? Can you tell me why O2PLS is much better than the others?

Kind regards,

Francesca

[Hengkov]

Hi Francesca,
SmartPLS not option NIPALS algorithm and others, just PLS Algorithm with three shceme (centroid, factorial and path).
But I know this fitur option (NIPALS Algorithm) in XLSTAT-PLS software, I am right?
For NIPALS algorithm, OPLS, and O2PLS is similar result (because used least squares estimation).
Regards,
Hengky