beer <- read.table("http://www.unt.edu/rss/class/mike/data/beer.txt",header=TRUE, sep="", na.strings="NA", dec=".", strip.white=TRUE)
beer2 = beer[,1:7] #just the attribute variables

#Installed from the Bioconductor repository. It is a relatively recent package and still under development itself, but given that it came out after this version
of R was released you may encounter issues

library(pcaMethods) #this package depends on the aroma.light package also

#Will estimate missing values with an algorithm related to that used for partial least squares regression, and is actually a nice tool to apply to any dataset for that reason alone
fit2=pca(beer2, method = "nipals",nPcs=3)  
fit2  #complete output
fit2@completeObs  #the now complete dataset, see the values on AROMA
summary(fit2)
round(fit2@loadings, 2)
slplot(fit2)
plotPcs(fit2)

#Seems to be unavailable to R 2.6.2 in the lab. Robust PCA, will give warning if missing values are present; alternatively one could first create a robust correlation matrix via cov.rob in the MASS(VR) package
fit3=pca(beer2, method = "robustPca",nPcs=3) 
fit3  #complete output
summary(fit3)
fit3@loadings 
round(fit3@loadings,2)  #Same, just rounded to two decimal places
slplot(fit3)  #score/loading plot
plotPcs(fit3)


#Methods available

method=c("svd", "nipals", "bpca", "ppca", "svdImpute", "nlpca", "robustPca")
svd: Uses classical prcomp. See documentation for svdPca.
nipals: An iterative method capable of handling small amounts of missing values. See documentation for nipalsPca.
bpca: An iterative method using a Bayesian model to handle missing values. See documentation for bpca.
ppca:  An iterative method using a probabilistic model to handle missing values. See documentation for ppca.
svdImpute: Uses expectation maximation to perform SVD PCA on incomplete data. See documentation for svdImpute.