wines: Piedmont wines data

Description Usage Format Details Source References Examples


Data refer to chemical properties of 178 specimens of three types of wine produced in the Piedmont region of Italy.




A data frame with 178 observations on the following 28 variables.

wine wine name (categorical, levels: Barbera, Barolo, Grignolino)
alcohol alcohol percentage (numeric)
sugar sugar-free extract (numeric)
acidity fixed acidity (numeric)
tartaric tartaric acid (numeric)
malic malic acid (numeric)
uronic uronic acids (numeric)
pH pH (numeric)
ash ash (numeric)
alcal_ash alcalinity of ash (numeric)
potassium potassium (numeric)
calcium calcium (numeric)
magnesium magnesium (numeric)
phosphate phosphate (numeric)
cloride chloride (numeric)
phenols total phenols (numeric)
flavanoids flavanoids (numeric)
nonflavanoids nonflavanoid phenols (numeric)
proanthocyanins proanthocyanins (numeric)
colour colour intensity (numeric)
hue hue (numeric)
OD_dw OD₂₈₀/OD₃₁₅ of diluted wines (numeric)
OD_fl OD₂₈₀/OD₃₁₅ of flavanoids (numeric)
glycerol glycerol (numeric)
butanediol 2,3-butanediol (numeric)
nitrogen total nitrogen (numeric)
proline proline (numeric)
methanol methanol (numeric)


The data represent 27 chemical measurements on each of 178 wine specimens belonging to three types of wine produced in the Piedmont region of Italy. The data have been presented and examined by Forina et al. (1986) and were freely accessible from the PARVUS web-site until it was active. These data or, more often, a subset of them are now available from various places, including some R packages. The present dataset includes all variables available on the PARVUS repository, which are the variables listed by Forina et al. (1986) with the exception of ‘Sulphate’. Moreover, it reveals the undocumented fact that the original dataset appears to include also the vintage year; see the final portion of the ‘Examples’ below.


Forina, M., Lanteri, S. Armanino, C., Casolino, C., Casale, M. and Oliveri, P. V-PARVUS 2008: an extendible package of programs for esplorative data analysis, classification and regression analysis. Dip. Chimica e Tecnologie Farmaceutiche ed Alimentari, Università di Genova, Italia. Web-site (not accessible as of 2014):


Forina M., Armanino C., Castino M. and Ubigli M. (1986). Multivariate data analysis as a discriminating method of the origin of wines. Vitis 25, 189–201.


pairs(wines[,c(2,3,16:18)], col=as.numeric(wines$wine))
code <- substr(rownames(wines), 1, 3)
table(wines$wine, code)
year <- as.numeric(substr(rownames(wines), 6, 7))
table(wines$wine, year)
# coincides with Table 1(a) of Forina et al. (1986)

sn documentation built on May 29, 2017, 7:18 p.m.

Search within the sn package
Search all R packages, documentation and source code