作者: Samuel D. Oman , Tormod Naes , Anan Zube
关键词:
摘要: A new regression method for non-linear near-infrared spectroscopic data is proposed. The technique based on a model which linear in the principal components and simple functions (squares products) of them. Added variable plots are used to determine squares products incorporate into model. coefficients estimated by Stein estimate shrinks towards determined first several selected terms. not computationally intensive appropriate routine predictions chemical concentrations. tested three sets all cases gives more accurate than does regression.