COVID-19 Update: We are currently shipping orders daily. However, due to transit disruptions in some geographies, deliveries may be delayed. To provide all customers with timely access to content, we are offering 50% off Science and Technology Print & eBook bundle options. Terms & conditions.
Data Analysis for Hyphenated Techniques - 1st Edition - ISBN: 9780444822376, 9780080544755

Data Analysis for Hyphenated Techniques, Volume 17

1st Edition

0.0 star rating Write a review
Authors: E.J. Karjalainen U.P. Karjalainen
eBook ISBN: 9780080544755
Imprint: Elsevier Science
Published Date: 14th March 1996
Page Count: 470
Sales tax will be calculated at check-out Price includes VAT/GST
Price includes VAT/GST

Institutional Subscription

Secure Checkout

Personal information is secured with SSL technology.

Free Shipping

Free global shipping
No minimum order.

Table of Contents

Part 1 - Analysis of Hyphenated Data 1. Analysis of overlapping spectra. 2. Data preprocessing. 3. Compression with principal components. 4. Techniques for library searches. 5. Neighborhood operations on hyphenated data. 6. Alternating regression. 7. Applying the OSCAR algorithm - A practical example. 8. Applications in other spectroscopies. 9. Validation. 10. AR and factor analysis. 11. Looking ahead.

Part 2 - Computer Programs for Hyphenated Data 12. The starting point. 13. Selecting and preprocessing the raw data. 14. Gathering the statistics from the AR experiment. 15. AR statistics in two dimensions. 16. AR statistics in three dimensions. 17. Selecting an optimal paramater set. 18. Displaying the spectra and elution curves. 19. Calculating the confidence ranges of spectra and elution curves. 20. Looking at the confidence rangs of the spectra and elution curves. 21. Looking at the measurements. 22. Finding help. Appendix. Index.


Based on a concrete set of working MATLAB programs, the book begins with a short program of snippets describing basic principles and proceeds to a complete application program filling a central analytical need: obtaining pure spectra from the observed overlapping spectra, with standard deviations for the solutions obtained.

The strength of the book is the open nature of the programs. All programs can be read, tested and modified by the user. The mathematical principles needed for the proper treatment of experimental data are thoroughly described. The first part of the book describes how data is collected, converted and prepared for the actual deconvolution. Practical working algorithms such as the Savitzky-Golay smoothing method are emphasized. The reader can see how searches in spectral libraries can be greatly speeded up by proper formulation of the calculation. Basic signal processing is described by illustrative examples.

The main part of the book describes the deconvolution of overlapping chromatographic peaks. Principle component analysis is described and used as a useful tool. The main emphasis is a discussion of a deconvolution method, OSCAR (Optimization by Stepwise Constraining of Alternating Regression), developed by the authors. The results can be validated as OSCAR calculates confidence intervals for the spectra and elution curves.

For users who do not want to enter the programs by hand a separate CD-ROM is available. It contains the programs and extensive sample data files. The CD-ROM has instructional multimedia showing step by step how the programs are used for the problems in the book. The MATLAB programs and data files can be directly run on Windows® and Macintosh® computers having a MATLAB interpreter.


No. of pages:
© Elsevier Science 1996
14th March 1996
Elsevier Science
eBook ISBN:

Ratings and Reviews

About the Authors

E.J. Karjalainen

U.P. Karjalainen

Affiliations and Expertise

Department of Clinical Chemistry, University of Helsinki, Finland