Applied Multivariate Analysis (348-0-20)
Instructors
Thomas A Severini
847.467.1254
2006 Sheridan Road/Room 305/Department of Statistics
Meeting Info
Technological Institute M164: Tues, Thurs 9:30AM - 10:50AM
Overview of class
The primary goal of this course is to present statistical methods for describing and analyzing
multivariate data, data in which the response is multidimensional. Topics covered will include principal component analysis, canonical correlation, multidimensional scaling, factor analysis, and clustering. Although the statistical theory behind the methods will be discussed, the course will emphasize the statistical and geometric motivation for the methods, the practical application of the methods, and the interpretation of the results. Given the nature of the data, and the methodology used, linear algebra plays an important role in the course. Thus, a second goal of the course is to present topics in linear algebra that are useful in understanding
statistical methodology.
Registration Requirements
STAT 320-2, STAT 350, and MATH 240. The course will use R extensively; hence, some experience with R will be useful.
Learning Objectives
1. To understand important properties of multivariate distributions;
2. To understand the statistical and geometric bases for multivariate statistical methods, including principal components analysis, canonical correlation analysis, factor analysis, multidimensional scaling and cluster analysis;
3. To be able to apply multivariate statistical methods using R and to be able to interpret the results
4. To understand the role of linear algebra in multivariate statistical methodology
Evaluation Method
Grades will be based on homework, an exam, and a final project, equally weighted.
Class Materials (Required)
Course notes, which will play the role of the course text, will be provided.
Class Materials (Suggested)
Students with limited experience with R may find the following text useful:
Introductory Statistics with R by P. Dalgaard, published by Springer (2008).
This book is available as an electronic resource on NUCAT/NUsearch.
Class Attributes
Formal Studies Distro Area