STA 345 Multivariate Statistical Analysis
Instructor : Woncheol Jang
Email: wjang@stat.duke.edu
Office : 223B Old Chem
Course syllabus:[pdf]
Lecture:
WF 10:05AM-11:20AM, Old Chem 116M
Required Text:
Johnson, R.A. and Wichern, D.A. (2002)
Applied
Multivariate Statistical Analysis. Fifth Edition. Prentice Hall.
Recommended Text:
Everitt, B. (2005).
An R and S-plus companion to multivariate
analysis [html]. Springer.
Hardle, W. and Simar, L. (2003).
Applied Multivariate Statistical Analysis [html]. Springer.
Prerequisites:
I assume you have taken STA 213 Introduction to Statistical
Methods and STA 244 Linear Models. In other words, you should
know linear algebra and statistical principles at
the level of Casella and Berger (2002) or Wasserman (2004).
Course Description:
This course covers methods for analyzing continuous multivariate
data. Broadly, we discuss
- Modeling and inference using the multivariate normal distribution
- Multivariate data and models
- Multivariate Normal distribution
- Traditional inference: Multivariate Regression, MANOVA, etc
- Links with mixed linear models and hierarchical modeling
- Exploratory techniques based eigenvalue and singular decomposition
- SVD of a data matrix; special decomposition.
- Principle Component Analysis
- Factor Analysis
- Canonical Correlation
- Classification and Clustering
- Linear Discrimination
- Classification Trees
- Hierarchical Clustering
- K-means Clustering
- Multidimensional Scaling
- Functional data analysis (if time permits)
- Functional PCA
- Functional Classification
- Functional Clustering
Computing:
Homework/Final Project:
Data:
Handout:
Lecture: