Finding Structures of Interest in a Large Data Set Using Factor Analysis

Abstract

In this paper we introduce a statistical method which can be used in combination with principal component analysis or factor analysis. Certain variables of a large data set which are of interest can be selected in order to calculate loadings and scores of these variables. We describe how the remaining variables of the data set can be presented in the previously extracted factor space. Furthermore, a possibility for the representation of the results is shown which is helpful for the interpretation.