Cross-validation for 10 groups

[SASdevAnneMarie]

Hello Experts,

I'm wondering if, for performing 10-fold cross-validation with the proc pls method=pcr method, I should use cv=split(10) cvtest(seed=12345)?

Thank you for your help!

[Ksharp]

I think "cv=split(10)" is good enough.

and I quoted:

CV=TESTSET(SAS-data-set)
specifies a test set of observations to be used for validation (formally, this is called “test set validation”
rather than “cross validation”).

or using "CV=ONE" or "CV=BLOCK(10)" 

[sbxkoenk]

"cvtest" is optional indeed (concurring with @Ksharp).

... to be added only if you want van der Voet’s (1994) randomization-based model comparison test be performed.

See documentation:

SAS® Viya® Platform Programming Documentation |  2026.04

SAS/STAT User's Guide

The PLS Procedure

Cross Validation (in "details" section)

https://go.documentation.sas.com/doc/en/pgmsascdc/v_074/statug/statug_pls_details07.htm

CV= Specifies the cross validation method to be used

https://go.documentation.sas.com/doc/en/pgmsascdc/v_074/statug/statug_pls_syntax01.htm#statug.pls.pl...

CVTEST Specifies that van der Voet’s (1994) randomization-based model comparison test be performed

https://go.documentation.sas.com/doc/en/pgmsascdc/v_074/statug/statug_pls_syntax01.htm#statug.pls.pl...

PROC PLS can specify various methods to be used for cross validation.

• PROC HPPLS (multi-threaded High Performance PLS) supports test set validation only by using the PARTITION statement.
• PROC PLSMOD (distributed computing = multi-node and multi-threaded in-memory CAS-enabled PLS) supports test set validation only by using the PARTITION statement.

BR, Koen