K. Faber et al., INTERRELATIONSHIPS BETWEEN SENSITIVITY AND SELECTIVITY MEASURES FOR SPECTROSCOPIC ANALYSIS - COMMENT, Chemometrics and intelligent laboratory systems, 38(1), 1997, pp. 89-93
In a recent paper Kalivas and Lang state that Lorber's multivariate se
nsitivity only relates to the unsealed variance inflation factor if th
e number of calibration samples equals the number of analytes i.e. if
pure-component spectra are used to form the response matrix. This pape
r gives a proof that Lorber's multivariate sensitivity obeys the desir
ed relationship under the general conditions implied by a good calibra
tion design. Emphasis is placed on making a clear distinction between
expected (true), observed (measured) and estimated quantities. This fa
cilitates the discussion of the practical evaluation of the defining e
quations and thereby solves some problems that are present in the orig
inal work. The alternative definition for multivariate sensitivity pro
posed by Kalivas and Lang is also discussed. (C) 1997 Elsevier Science
B.V.