# read training data 
con = url("http://www.tau.ac.il/~saharon/BigData2015/covtrain.csv")
train = read.csv(con)

# reorganize nicely
p=dim(train)[2]
n = dim(train)[1]
X = as.matrix(train[,1:p],nrow=n)

# Use the functions eigen() and/or prcomp() - read their help carefully and make sure you interpret the results correctly. 
# Ask yourself: Are PC loadings in rows or columns? Are eigen values or singular values (their square root) reported? Does our model require centering?  
# How do we make sure the function handles the centering aspect as we want it?  
?eigen
?prcomp

# For plotting you can use hist() or better plot(density(...))
# You can add lines to the plot using lines()
# Make sure you scale the plots properly: 
# Is y axis density or counts? Both are legitimate but the different types of information on the same graph have to be on the same scale