710 P.J. Mart ́ın-Alvarez ́
–5 0 5 10 15 20 25 30 35
Observed values of aging time (months)–505101520253035Predicted values of aging time (months)Fig. 13.7Predicted values obtained using PLS regression vs observed values for aging time
vs the observed values for time. As can be seen, the fit for the predictions of aging
time can be considered to be appropriate.
We have also used these techniques in the area of dairy products, more specifi-
cally to determine the percentage composition of milk blends (Molina et al. 1995,
1999; Recio et al. 2004), to predict the ripening time of cheeses (Poveda et al. 2004;
Garc ́ıa-Ru ́ız 1998), and to predict sensory attributes of cheeses (Cabezas et al. 2006;
Gonz ́alez-Vi ̃nas et al. 2007).
AcknowledgmentsThis work was supported by grants AGL2006-04514 and AGL2005-03381,
from the Spanish Ministerio de Educaci ́on y Ciencia.
References
Afifi, A.A., & Azen, S.P. (1979). Statistical analysis. A computer oriented. Academic Press, Inc.
New York.
Afifi A.A., & Clark V. (1996). Computer-Aided Multivariate Analysis. Chapman & Hall, London.
Alcaide-Hidalgo, J.M., Moreno-Arribas, M.V., Mart ́ın-Alvarez, P.J., & Polo, M.C. (2007). Influ- ́
ence of malolactic fermentation, postfermentative treatments and ageing with lees on nitrogen
compounds of red wines.Food Chem., 103, 572–581.
Barcenilla, J.M., Martin-Alvarez, P.J., Vian, A., & Gonz ́alez, R. (2003). Characterization of yeast
strains on the basis of autolytic activity and volatile compounds.Food Sci. Technol. Int., 9,