ABSTRACT
The main aim of principal components analysis (PCA) is to replace p metrical correlated variables by a much smaller number of uncorrelated variables which contain most of the information in the original set. This greatly simplifies the task of understanding the structure of the data since it is much easier to interpret two or three uncorrelated variables than 20 or 30 that have a complicated pattern of interrelationships. In order to translate this objective into a practical method, we have to be more precise about what it is to retain “most of the information”.
