A review of variable selection methods in Partial Least Squares Regression

With the increasing ease of measuring multiple variables per object the importance of variable selection for data reduction and for improved interpretability is gaining importance. There are numerous suggested methods for variable selection in the literature of data analysis and statistics, and it is a challenge to stay updated on all the possibilities. We therefore present a review of available methods for variable selection within one of the many modeling approaches for high-throughput data, Partial Least Squares Regression. The aim of this paper is mainly to collect and shortly present the methods in such a way that the reader easily can get an understanding of the characteristics of the methods and to get a basis for selecting an appropriate method for own use. For each method we also give references to its use in the literature for further reading, and also to software availability.