Complete course on Exploratory Multivariate Data Analysis (MOOC)
This course corresponds to a MOOC (Massive Open Online Course) that is free and played in February or March (the first session was in 2017). You can subscribe From December to March from the platform FUN.
The links in blue correspond to the course videos or the course slides on the method, the links in brown correspond to tutorial's on FactoMineR or on missing values.
Note that sometimes we refer to quiz and exercises but they are only available on the MOOC.
Introduction
1. Principal Component Analysis (PCA)
- Data - practicalities
- Studying individuals and variables
- Interpretation aids
- PCA with FactoMineR
- Factoshiny: interactive graphs in exploratory multivariate data analysis
- Handling missing values in PCA
- Slides on the PCA course
- Audio transcription of the PCA course
- A detailed PCA example
- Wine dataset (course) and R code; Decathlon dataset (software) and R code
2. Correspondence Analysis (CA)
- Introduction
- Visualizing the row and column clouds
- Inertia and percentage of inertia
- Simultaneous representation
- Interpretation aids
- CA with FactoMineR
- Text mining with correspondence analysis
- Slides on the CA course
- Audio transcription of the Correspondence Analysis course
- A detailed CA example
- Nobel dataset (course) and R code; Birth dataset (software) and R code
3. Multiple Correspondence Analysis (MCA)
- Data - issues
- Visualizing the point cloud of individuals
- Visualizing the cloud of categories
- Interpretation aids
- MCA with FactoMineR
- Handling missing values in MCA
- Slides on the MCA source
- Audio transcription of the Multiple Correspondence Analysis course
- A detailed MCA example
- Hobbies dataset (course) and R code; Tea dataset (software) and R code
4. Clustering
- Introduction
- Example and how to choose the number of clusters
- The partitioning method K-means
- Characterizing clusters
- Clustering with FactoMineR
- Slides on the clustering course
- Audio transcription of the Clustering course
- A detailed clustering example
- Temperature dataset (course) and R code; Decathlon dataset (software) and R code
5. Multiple Factor Analysis
- Introduction
- Weighting and global PCA
- Study of the groups of variables
- Complements: qualitative groups, frenquency tables
- MFA with FactoMineR
- Slides on the MFA course
- Audio transcription of the Multiple Factor Analysis course
- A detailed MFA example
- Wine juries dataset (course) and R code; Sensory dataset (available in FactoMineR) and R code