What is it about?
Among the statistical methods used to define a class space UNEQ is one of greater importance. It was defined by Hotelling in 1931, and used in multivariate quality control, i.e. when the quality is controlled by means of many parameters. This paper describes the correct statistical procedure to be used with UNEQ. The class space has (when the number of control parameters is only 2) the form of an ellipse in the space of the two parameters. Samples within the suitable ellipse have a sufficient quality.
Featured Image
Why is it important?
Because of the large use of UNEQ in quality control the improvement in the associated statistics is very important.
Perspectives
In the past years the use of multivariate quality control increased very much, and it will be further increase because of the imprtance of quality control, origin control, process control,, and so on. An example in food quality control: the virgin olive oil can have from 63 to 88% oleic acid, and from 6 to 18% palmitic acid. The extreme values are not probeble, so that samples can be suspected of adulteration. The correlation between the two acids is negative and very large: it means that is not possible that an authetic sample contains bot very large amount of the two acids, or very small content . When this strategy is enlarge to include all fat acids in the oil, sterols, and minor components, the fraud becomes almost impossible .
Professor Michele Forina
University of genova
Read the Original
This page is a summary of: Distance and class space in the UNEQ class-modeling technique, Journal of Chemometrics, March 1995, Wiley,
DOI: 10.1002/cem.1180090202.
You can read the full text:
Contributors
The following have contributed to this page
