Aunir Glossary - PLS

Aunir glossary for July

Back again for another installment, here is the Aunir glossary. This month we look at PLS... 

PLS stands for Partial Least Square and is a type of regression analysis. It is an analytical technique which is a form of data compression. WIth PLS the rule used to derive the factors is that each factor in turn maximises the covariance between the Y data and all possible linear combinations of the X data. PLS is therefore a balance between variance and correlation with each factor being influenced by both effects. PLS factors are therefore more directly related to variability in Y values than are Principal Components. PLS produces three new variables, loading weights (which are not orthogonal to each other), loadings, and scores which are both orthogonal. PLS models are produced by regressing PLS scores against Y values. The regression coefficients in PLS space are usually converted back to a prediction model using all the data points in wavelength space.


This definition is adapted from the glossary of 'Quality Assurance for Animal Feed Analysis Laboratories' which was co-authored by Jim Balthrop, Benedikt Brand, Richard A. Cowie, Jürgen Danier, Johan De Boever, Leon de Jonge, Felicity Jackson, Harinder P.S. Makkar and not forgetting our own Chris Piotrowski.