MULTIVAR: a program for multivariate calibration incorporating net analyte signal calculations. H. Goicoechea y A. Olivieri

GOICOECHEA, HÉCTOR C; AC OLIVIERI,

TRAC-TRENDS IN ANALYTICAL CHEMISTRY

Elsevier

Año: 2000 vol. 19 p. 599 - 605

0165-9936

A useful chemometric tool is the use of multivariate calibration applied to spectroscopic data [ 1], which enables the ef¢cient extraction of information concerning certain analytes of interest from spectra of multicomponent mixtures. Two popular methods are principal component regression (PCR) and partial least-squares (PLS) [ 2 ], both of which use inverse calibration steps combined with a prior optimisation of the calibration information. They have the following advantages: (1) use of full spectra, (2) knowledge of only the concentrations of the analytes of interest in the calibration samples is required, and (3) spectral decomposition into factors avoids the problems associated with spectral collinearities [ 2 ]. They are ideally suited for the study of complex biological samples, as in drug or metabolite monitoring in blood [ 3,4 ], or in pharmaceutical analysis of certain multicomponent preparations where not all the excipients may be known [ 5 ]. Acommon requirement of all thesemultivariate methods is that the background should be modelled during calibration, i.e. compounds of no interest ( but present in unknown samples) should be contained in the calibration samples.