STAT 832

STAT 832: Multivariate Statistical Analysis

Course information

• Instructor: Peter Hoff
• TA: Bongjung Sung
• Lecture: Tue, Thu 3:05-4:20, Old Chemistry 025
• Office hours: Mon, Wed 9:00-10:00 AM Old Chem 219 (PH), Thu 4:30-6:30 PM Old Chem 025 (BS)

---

Course materials

• Lecture notes, code and data
  
• Canvas site
  
• Multivariate Analysis (Mardia, Kent and Bibby)
• Matrix-Based Introduction to Multivariate Data Analysis (Adachi)
• Applied Multivariate Statistical Analysis (Hardle and Simar)
• Modern Multivariate Statistical Techniques (Izenman)

---

Tentative schedule of topics

1. Unsupervised analysis
   1. Matrix decomposition and PCA
   2. Mean and covariance estimation and testing
   3. Factor analysis
2. Supervised analysis
   1. General linear model
   2. Classification
3. Special topics (if time)
   1. Kernel methods
   2. Multilinear models
   3. Copulas
   4. Clustering

---

Evaluation

• 70% Homework
• 30% project, due 2026/04/28
• 10% per-late day penalty (including weekends)