Partial Least Square (PLS)
PLS is simply a derivation of the multiple linear regression and principal components regression methods. PLS generates new variables called principal components which are linearly transformed from a group of high correlated variables into a smaller number of uncorrelated variables that are able to explain the variability observed in the data. Subsequently the newly generated principal components are used for the prediction of the dependent variable.
- Eriksson L, Johnansson E, Wold S (2001) Basic concepts and principles of projection. In: multi- and megavariate data analysis. Principles and applications, Umetrics Academy, Umea, pp 21–41Google Scholar
- Geladi P, Kowlaski B (1986) Partial least square regression: a tutorial. Anal Chem Acta 35:1–17Google Scholar