This book is the first comprehensive account of the multiple factor analysis (MFA) in English. This method can be considered as an extension of both principal component analysis and multiple correspondence analysis: it is devoted to exploring tables in which a group of individuals is described by several sets of variables, these variables being quantitative and / or qualitative. Its scope is very wide as shown by its many uses in various fields (sensory analysis, investigations, genomics, etc.).
This book adopts 3 principles:
- Special emphasis is given to examples to illustrate the questions to answer, the properties of the methods and the rules for interpreting their results.
- Geometric reasoning is favoured in order to make the presentation as non-technical as possible; matrix formalisations are explained subsequently and only when necessary.
- Each chapter ends with a description of the implementation of methodologies presented with the free R package FactoMineR.
Features: Many examples that illustrate the properties of MFA and the rules for interpreting the results Use of geometric reasoning to simply describe the properties of the MFA Mathematical properties extensively covered to facilitate intuitive understanding Description of implementation with FactoMineR. The data presented and the commented R commands are available here.