I am trying to figure out the code for computing principal components using princomp procedure with a covariance or correlation matrix as an input. I have been able to find the solution using raw data (through covariance or correlation). But now I dont have the raw data and have the covariance matrix instead.