Abstract

Dimensionality reduction techniques such as Principal component /Factor analysis and correspondence analysis maybe limited in the presence of data that are grouped under well-defined themes or when different variable types are involved (continuous and categorical) Multiple Factor Analysis (MFA) is dedicated to datasets where variables are structured into groups and when several sets of variables (continuous or categorical) are simultaneously studied. In this seminar, you are going to explore the basic features of MFA using the FactoMineR package which further allows you to generate clusters using the principal components. Attention will be given to useful outputs and their interpretations.