Analysis of mixture data with partial least squares

https://doi.org/10.1016/0169-7439(92)80092-IGet rights and content

Abstract

Kettaneh-Wold, N., 1992. Analysis of mixture data with PLS. Chemometrics and Intelligent Laboratory Systems, 14: 57–69.

The analysis of mixture data is a common problem in industrial research and development, particularly in chemical and related industries, e.g. pharmaceuticals, cosmetics, oil, and biotechnology. Analyzing mixture data with multiple regression necessitates special model forms due to the mixture constraint. The canonical polynomials of Scheffé and of Cox will be discussed, as well as the limitation of multiple regression with data in constrained regions. For the analysis of mixture data, partial least squares (PLS) has been found to be practical. In particular when both mixture and process variables are involved, it offers a flexible and simple approach which works well in practice. The analysis of mixture data using PLS and multiple regression are compared, with case studies from the scientific literature.

References (9)

  • J.A. Cornell
  • D.R. Cox

    A note on polynomial response functions for mixtures

    Biometrica

    (1971)
  • H. Wold

    Soft modeling. The basic design and some extensions

  • S. Wold et al.

    Multivariate data analysis in chemistry

There are more references available in the full text version of this article.

Cited by (94)

  • Experimental design application and interpretation in pharmaceutical technology

    2023, Computer-Aided Applications in Pharmaceutical Technology: Delivery Systems, Dosage Forms, and Pharmaceutical Unit Operations
  • Optimization of the formulation of an original hydrogel-based bone cement using a mixture design

    2020, Journal of the Mechanical Behavior of Biomedical Materials
  • A comprehensive strategy in the development of a cyclodextrin-modified microemulsion electrokinetic chromatographic method for the assay of diclofenac and its impurities: Mixture-process variable experiments and quality by design

    2016, Journal of Chromatography A
    Citation Excerpt :

    The set-up of the experimental matrices and the related statistical treatment of the data was made by MODDE 10 [62] software. Projection to Latent Structures regression (PLS) was employed as regression technique [63,64]. The definition of DS was based on the obtained PLS models and Monte-Carlo simulations [59], by taking into account the error in prediction of the models.

View all citing articles on Scopus
View full text