Chemometrics and Intelligent Laboratory Systems vol:149 pages:50-59
Sparse partial robust M regression is introduced as a new regression method. It is the ﬁrst dimension reduction and regression algorithm that yields esti-mates with a partial least squares like interpretability that are sparse and robust with respect to both vertical outliers and leverage points. A simula-tion study underpins these claims. Real data examples illustrate the validity of the approach.