Functional data analysis for the development of a calibration model for near-infrared data

Jiang, C; Martin, EB

doi:10.1016/S1570-7946(08)80119-8

Functional data analysis for the development of a calibration model for near-infrared data

Lookup NU author(s): Cheng Jiang, Professor Elaine Martin

Downloads

Full text for this publication is not currently held within this repository. Alternative links are provided below where available.

Abstract

The calibration performance of two functional data analysis approaches is investigated in this paper. The performance of the method is considered for different basis functions, the penalized B-spline with equally spaced knots and the B-spline with unequally spaced knots and wavelets. The various approaches are compared with respect to the prediction of the mole fractions of different components from the spectrum of mixture samples. The different approaches are benchmarked against the more traditional calibration modelling approach of PLS. © 2008 Elsevier B.V. All rights reserved.