On orthogonal signal correction教程.pdfVIP

Chemometrics and Intelligent Laboratory Systems 50 2000 47–52 locatechemometrics On orthogonal signal correction Tom Fearn Department of Statistical Science, Uniersity College London, Gower Street, London WC1E 6BT, UK Received 1 June 1999; accepted 3 July 1999 Abstract A new algorithm for orthogonal signal correction is presented, compared with existing algorithms, and illustrated on an example from near infrared spectroscopy. Given a matrix X of spectral or other high dimensional data and a vector or matrix Y of concentrations or other reference measurements on the same samples, orthogonal signal correction subtracts from X factors that account for as much as possible of the variance in X and are orthogonal to Y. The aim is to improve the perfor- mance of a subsequent partial least squares PLS regression of Y on X. 2000 Elsevier Science B.V. All rights reserved. Keywords: Chemometrics; Near infrared; Spectroscopy; Orthogonal signal correction; Partial least squares 1. Introduction large, and will typically exceed n. The aim is to pro- duce a calibration equation to predict concentrations Orthogonal signal correction OSC was intro- from spectra for future samples using, for example, duced by Wold et al. 1 and applied to calibration partial least squares PLS regression. transfer by Sjoblom et