STAT 142 - Applied Multivariate Analysis
Course Description
Principal component analysis; factor analysis; discriminant analysis; cluster analysis; other multivariate techniques
Course Learning Outcomes
After completion of the course, the student should be able to:
- Demonstrate understanding of the basic concepts, theories and methodologies of multivariate analysis;
- Conduct appropriate multivariate statistical analyses of data; and
- Interpret correctly the results for application.
Course Outline
Unit 0. Preliminaries and Introduction
- Introduction to the Course
- Applications of Multivariate Analysis
Unit 1. Principal Component Analysis
- Definition
- Dimension Reduction Properties
- Population Principal Components
- Sample Principal Components
- Inference on Sample Component
- Applications
Unit 2. Factor Analysis
- Underlying Model
- Estimation Procedures
- Applications
Unit 3. Discriminant Analysis
- Introduction
- Two Groups: Known Distributions
- Two Groups: Known Distributions with
Unknown Parameters
- Two Groups: Unknown Distributions
- More Than Two Groups
- Applications
Unit 4. Cluster Analysis
- Similarity Measures
- Hierarchical Clustering: Agglomerative
Technique
- Hierarchical Clustering: Divisive Techniques
- Partitioning Methods
- Choosing the Number of Clusters
- Applications
Unit 5. Other Multivariate Techniques
- Canonical Correlation Analysis
- Multidimensional Scaling
- Procrustes Analysis
- Applications