Permutation test as possible alternative to the commonly used cross-validation of samples for validating PCA results is suggested. The approach is then illustrated using two data-sets from consumer studies of apple and raspberry juice. Our findings show that internal validation provided by the permutation test is particularly advantageous when the data are complex as they are in the second case reported here.
|Citation:||Endrizzi, I.; Gasperi, F.; Rodbotten, M.; Naes, T. (2013). Permutation testing for validating PCA. In: Advances in Latent Variables: Methods, Models and Applications, Brescia, June 19–21, 2013. ISBN: 978-88-343-2556-8. url: http://meetings.sis-statistica.org/index.php/sis2013/ALV/paper/viewFile/2632/409 handle: http://hdl.handle.net/10449/22414|
|Organization unit:||Food Quality and Nutrition Department # CRI_2011-JAN2016|
|Authors:||Endrizzi, I.; Gasperi, F.; Rodbotten, M.; Naes, T.|
|Title:||Permutation testing for validating PCA|
|Scientific Disciplinary Area:||Settore SECS-S/02 - Statistica Per La Ricerca Sperimentale E Tecnologica|
|Appears in Collections:||03 - Conference object|