Principal Component Analysis is a statistical procedure that uses an orthogonal transformation to convert a set of observations of possible correlated variables into a set of values of linear uncorrelated variables called principal components. The goal is to explain the maximum amount of variance with the fewest number of principal components. PCA transforms the initial features into new ones that are linear combinations of the set of variables. For this analysis, first the original values should be normalized and the covariance matrix should be formed. The eigenvalues and eigenvectors should be calculated and the eigenvector with the highest eigenvalue has to be chosen. For the highest eigenvalue the data set matrix has to be multiplied and finally the mean can be put back which was removed in the beginning. However, if the original data set is correlated the solution can be unstable. Read more at: http://www.datasciencecentral.com/profiles/blogs/introduction-to-principal-component-analysis