Course outline   

This is an introductory multivariate statistical analysis course designed for graduate students in the Department of Statistics. The aim of the course is to introduce a variety of standard statistical methods used to analyze multivariate data, emphasizing the implementation and interpretations of these methods. In this course, the topics to be covered include:

• Graphical methods for Multivariate Data

• Multivariate Normality

• Multivariate ANOVA

• Principal Components

• Multidimensional Scaling

• Cluster Analysis

• Discriminant Analysis

• Factor Analysis

• Covariance Structure Models

• Canonical Correlation Analysis

    The course will include lecture and lab. The course will be given on the basis of 3 hours a week (Thursday, 2~5 PM) in 綜合三館 room 834. The purposes of lab are for students to reinforce ideas learned in the lecture and to practice the skills of real-data analysis. Success in both lecture and lab are required to pass the course. The materials for teaching will be regularly posted on the course website (http://www.stat.nthu.edu.tw/~swcheng/Teaching/stat5191/index.html). 

Computing

    The software I will be using for the course is R (S-plus is similar and will also work but is not free). R is free with Windows and Unix versions. You can download your own copy from its website and use it wherever you find convenient. 

Textbook

    Johnson, R.A. and Wichern, D.W. (2007), Applied Multivariate Statistical Analysis, 6th edition, Prentice Hall.

Reference

    Everitt B. (2005), An R and S-PLUS Companion to Multivariate Analysis, Springer. (Electronic version is available in library)

Assignment

    There will be assignments for each topics. You are free to discuss the assignment questions with your classmates, but you must write up your answers individually. You CANNOT turn answers in as a group or simply reproduce a classmate's write up. Please DO NOT hand in your assignments in the format of including all details of R outputs. You should present your results in summary format only with R outputs that are required to support your answers. All homework must be handed in before the end of class on the day it is due.

Contact method and office hours

    My office is in 綜合三館 room 818 (phone extension: 33162). My office hour is scheduled on Thursday, 5~6 PM in my office. Questions by e-mail are also welcome: my e-mail address swcheng@stat.nthu.edu.tw.

    The information of teaching assistants will be soon posted:

Grading

    Your grade will be determined by homework (30 %), a midterm (30 %), and a final exam (40 %). 

Prerequisites

    Knowledge of matrix algebra. Knowledge of basic probability and mathematical statistics. Computing will be required, but no specific prior experience is necessary.