Developing PLS algorithm: Utilizing PCA-power iteration-based weighting in R

AIP conference proceedings 2026

View at publisher
Authors: Angga Dwi Mulyanto, Bambang Widjanarko Otok, Jerry Dwi Trijoyo Purnomo
DOI: 10.1063/5.0321845
Cited by: 0
References: 10

Abstract

Incorporating Principal Component Analysis (PCA) with Power Iteration to the algorithm’s weighting mechanism this study develops a Partial Least Squares (PLS) Algorithm. To test the proposed algorithm, which was implemented in the R programming language, a case study of the effect of Organizational Culture on Lecturer Performance is used. Results reveal that the algorithm yields acceptable AVE and CR values showing adequate reliability and validity, and also that it produces consistent outer weights. The R² value is rather small but this new method provides a promising alternative to extend PLS-SEM for refining the calculation of outer weights.

Fields of Study & Subjects

Statistical Methods and Applications Advanced Statistical Methods and Models Psychometric Methodologies and Testing

Publication details

Journal / source
AIP conference proceedings
Volume
3416
Pages
040006–040006
Publication date
1 January 2026
Language
EN