Links

F. Husson website





Book Exploratory Multivariate Analysis Using R

Outline

Introduction

Principal Component Analysis

Correspondence Analysis

Multiple Correspondence Analysis

Clustering

Multiple Factor Analysis

To conclude

Forum

Exercise.

To answer questions in this exercise, it is not necessary to run a MCA. All that is needed is a good understanding of the principal properties of MCA. Each question is independent of the others.

We have questioned 8 people as to their socio-professional category (SPC) as well as that of their mother and father. We obtained the following data. We are going to think about doing MCA here.

MCA data set

Q1) We perform MCA on this table. What is the total inertia of the point cloud of individuals?
1
2
3
9

Q2) The 1st MCA axis is such that the projected inertia of the point cloud of the rows is equal to 0.833. What is the projected inertia of the point cloud of categories on the 1st axis?
0.833 * 9 / 3 = 2.5
0.833
on ne sait pas

Q3) In the category space, which category is closest to the origin? (give the category's label)
Fwhite-collar
Flaborer
Fprofessional
Mwhite-collar
Mlaborer
Mprofessional

Q4) In the category space, which category has the highest inertia?
Fwhite-collar
Flaborer
Fprofessional
white-collar
Mlaborer
Mprofessional

Q5) Here are the coordinate values of the individuals in the first 3 dimensions (rounded to 2 decimal places)
Resultat exo ACM
By applying the barycentric property, deduce the coordinates of the FWhite-collar category in the 1st dimension (Question 2 may be useful to help answer this).

-0.4
-0.54
-0.833
0.833
2

Score =
Correct answers: