Determination of chemical rank of two-way data from mixtures using subspace comparisons

A novel approach based upon subspace comparison is proposed for the determination of chemical rank of two-way data from mixtures. Pairs of subspaces spanned by Schmidt orthogonalized key variables from different methods are constructed. The chemical rank is determined as the maximum dimension for which the subspaces are in agreement with each other. Applications are made to the standard data sets made available by Windig and to a flow injection analysis data set by Gemperline and Hamilton.