# Classical Methods

When individuals are described by one set of variables, several methods are available depending on the types of variables considered (numerical or categorical variables):

- When variables are numericals one can perform a PCA (Principal components analysis).
- With a contingency table, one can perform a CA (Correspondence Analysis).
- When individuals are described by a set of categorical variables one can perform a MCA (Multiple Correspondence Analysis).

The HCPC (Hierarchical Clustering on Principal Components) function allows to perform a clustering on individuals. This function combines principal components methods, hierarchical clustering and partitioning to better visualize and highlight the similarities between individuals.

# Advanced Methods

## One set of individuals, several sets of variables

When individuals are described by several sets of variables, several types of analyses are proposed:

- MFA (Multiple Factor Analysis), for which the variables of a same group may be numerical or categorical or is a contingency table.
- HMFA (Hierarchical Multiple Factorial Analysis), an extension of MFA for which variables are structured according to a hierarchy.
- GPA (Generalized Procustean Analysis), for which variables must be continuous.

## One set of variables, several sets of individuals

When several sets of individuals are described by one set of continuous variables, the analysis proposed is an extension of MFA called Dual MFA.

## One set of individuals, two types of variables

When one set of individuals is described by one set of variables that may be continuous and/or categorical, the analysis proposed is an particular case of MFA called Factor Analysis of Mixed Data.

# Facto's best

You will find here some **FactoMineR**'s very specific functions:

*catdes()*for categories description*dimdesc()*for dimension description*condes()*for Continuous variables descriptions*plotellipses()*for confidence ellipses around categories after PCA or MCA- other useful functions