Standard Article

Multivariate Calibration of Analytical Data

Chemometrics

  1. Svante Wold1,
  2. Mats Josefson2

Published Online: 15 SEP 2006

DOI: 10.1002/9780470027318.a5205

Encyclopedia of Analytical Chemistry

Encyclopedia of Analytical Chemistry

How to Cite

Wold, S. and Josefson, M. 2006. Multivariate Calibration of Analytical Data. Encyclopedia of Analytical Chemistry. .

Author Information

  1. 1

    Umeå University, Research Group for Chemometrics, Sweden

  2. 2

    AstraZeneca R&D, Mölndal, Sweden

Publication History

  1. Published Online: 15 SEP 2006

Abstract

Multivariate calibration (MVC) is a methodology for using multiple signals, for instance a digitized spectrum, to determine the levels of concentrations of chemical compounds in analytical samples. MVC can also be used to determine other properties of interest, for instance viscosity, particle size distribution, energy content, or taste. MVC is made in two phases. In the first, the “training” or “calibration” phase, samples with known concentration (property) values and their signal profiles are used to develop a model of their relationship, a multivariate standard curve. In the second phase, this model is used with new samples to determine their concentration (property) values from their signal profiles.