Prediction of the storage time in bottles of Spanish white wines using multivariate statistical analysis

Abstract

 Young white wines lose their characteristic fruity aroma and flavour with storage time in the bottle. The present study employed three regression methods: stepwise multiple linear regression, principal component regression, and partial least squares regression, to predict the storage times of wines in the bottle based on the composition of their volatiles. Stepwise multiple linear regression yielded the best predictions (mean error by cross validation: 0.34 years), using ten volatile compounds in the equation (diethyl succinate, butyric acid, furan derivative, N–3-(methylbutyl) acetamide, ethyl lactate, furfural, diethyl malate, ethyl caprylate, ethyl-3-hydroxybutyrate, and ethyl-4-hydroxybutyrate).