Premium
Calculation of maximum and minimum band boundaries of feasible solutions for species profiles obtained by multivariate curve resolution
Author(s) -
Tauler R.
Publication year - 2001
Publication title -
journal of chemometrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.47
H-Index - 92
eISSN - 1099-128X
pISSN - 0886-9383
DOI - 10.1002/cem.654
Subject(s) - unimodality , normalization (sociology) , multivariate statistics , mathematics , rank (graph theory) , resolution (logic) , function (biology) , curve fitting , algorithm , computer science , statistics , combinatorics , artificial intelligence , evolutionary biology , sociology , anthropology , biology
A method for the calculation of maximum and minimum band boundaries of feasible solutions corresponding to the species profiles estimated by multivariate curve resolution is presented. The method is based on a non‐linear constrained optimization of an objective function defined by the ratio of the signal contribution of a particular species to the whole measured signal. Implementation of constraints such as normalization, closure, non‐negativity, unimodality and local rank/selectivity during the non‐linear optimization and their effect on the calculation of the feasible solutions are studied in detail. Calculation of the band boundaries is shown for different simulated and real two‐way data examples of increasing complexity. Copyright © 2001 John Wiley & Sons, Ltd.