|
Mahalanobis and Euclidean distances Tom
Fearn
Department of Statistical Science, University College London, Gower Street, London WC1E 6BT, UK
ABSTRACT:
This issue's Chemometrics space was sparked by two
independent recent queries about the use of Mahalanobis distance for detecting outliers. It discusses how, for NIR data, the Euclidean distance is preference, but the need for
dimension reduction complicates matters.
Keywords:
Back to Table of Contents |
RSS Feed
