Comparison of Linear Statistical Methods for Calibration of Nir Instruments

T. Naes, C. Irgens and H. Martens

Journal of the Royal Statistical Society Series C, 1986, vol. 35, issue 2, 195-206

Abstract: The multiple linear regression, ridge regression, principal component regression and partial least squares regression approach to statistical calibration of near infrared (NIR) instruments are compared. Computations on wheat data show that when the ratio between the number of calibration samples and the number of wavelengths in the NIR spectrum is low, the latter three methods, which are biased regression methods, give much better prediction results than multiple linear regression. This is very important in NIR analysis where this ratio is often small. In addition, we consider a new transformation of NlR data. It is shown that in company with the partial least squares method the transformation leads to very good results.

Date: 1986
References: Add references at CitEc
Citations:

Downloads: (external link)
https://doi.org/10.2307/2347270

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:bla:jorssc:v:35:y:1986:i:2:p:195-206

Ordering information: This journal article can be ordered from
http://ordering.onli ... 1111/(ISSN)1467-9876

Access Statistics for this article

Journal of the Royal Statistical Society Series C is currently edited by R. Chandler and P. W. F. Smith

More articles in Journal of the Royal Statistical Society Series C from Royal Statistical Society Contact information at EDIRC.
Bibliographic data for series maintained by Wiley Content Delivery ().