Course STAT 7331.

501 Multivariate Analysis

Professor Robert Serfling
Term Spring 2011
Meetings MW 5:30-6:45 pm, FO 2.608

Professor’s Contact Information

Office Phone 972-883-2361
Office Location FO 2.602C
Email Address serfling@utdallas.edu
Website www.utdallas.edu/~serfling
Office Hours TBA
Other Information I check my email much more regularly than my telephone.

General Course Information

Prerequisite STAT 6331 or the equivalent is required. This is firm.
Tentative Topics to be Covered: Curse of Dimensionality.
Geometry of Higher Dimensions. Dimension Reduction. Matrices
and Quadratic Forms. Best Linear Prediction. Multiple and Partial
Correlation. Multivariate Normal Distribution. Hotelling’s T2
Statistic. Inference for Means and Covariance Matrices. The
Course Wishart Distribution. Two- and Multi-Sample Problems.
Description MANOVA. The Classification Problem. Principal Components.
Factor Analysis. Testing Independence of Sets of Variates.
Multivariate Depth, Quantile, Sign, and Rank Functions.
Nonparametric Descriptive Measures for Multivariate Symmetry,
Location, Dispersion, Skewness, and Kurtosis. Equivariance and
Invariance Properties. Robustness Issues.
A working knowledge of multivariate statistical methods, taking
Desired Learning account of the interrelations among jointly distributed variates and
Outcomes exploiting geometrical aspects.
No text will be specified as required. Lectures will be from the
Required Text professor’s notes accompanied by handouts. Readings from various
books and journal articles will be assigned or recommended.
 Syllabus (subject to revision – except schedule for tests)
M 1/10 Introduction.
W 1/12 The Multivariate Normal Distribution.
M 1/17 Martin Luther King holiday – University closed
W 1/19 Graphical Display of Multivariate Data Clouds and Multivariate Functionals.
M 1/24 Estimation of Mean Vector and Covariance Matrix.
W 1/26 Distributions and Uses of Sample Correlation Coefficients.
M 1/31 Hotelling’s T2 Statistic.
W 2/2 Classification of Observations.
M 2/7 Sample Covariance Matrix and Related Statistics.
W 2/9 The General Linear Hypothesis. Multivariate ANOVA.
M 2/14 Factor Analysis and Related Methods.
W 2/16 Testing Independence of Sets of Variates.
M 2/21 Miscellaneous Results on Matrices and Quadratic Forms.
W 2/23 Principal Components.
M 2/28 Canonical Correlations and Canonical Variables.
W 3/2 Midterm Test (closed-book)
M 3/7 Distributions of Characteristic Roots and Vectors.
W 3/9 Curse of Dimensionality. Geometry of Higher Dimensions.
3/14-3/19 Spring Break ☺ – University closed
M 3/21 Dimension Reduction.
W 3/23 Representation and Geometry of Multivariate Data.
M 3/28 Multidimensional Classification and Clustering.
W 3/30 Nonparametric Description of Multivariate Distributions.
M 4/4 Multivariate Symmetry and Asymmetry.
W 4/6 Multivariate Depth Functions and Order Statistics.
M 4/11 Multivariate Sign and Rank Functions.
W 4/13 Multivariate Quantile Functions and Outlyingness Functions.
M 4/18 Multivariate Description via Generalized Univariate Quantile Functions.
W 4/20 Level Set Methods.
M 4/25 Student Presentations
W 4/27 Student Presentations
M 5/2 Student Presentations
Course Policies
The course grade will be based on homework (35%), a closed-book
midterm test (30%), and presentations (35%). The course will conclude
Grading
with preparation, under supervision of the instructor, of a 15-minute
Criteria
presentation to the class on an assigned research paper. The presentations
will be given together as a ``mini-symposium”.
In the case of a missed midterm test, a makeup exam (either written or
Missed Exam or
oral) will be conducted if the absence is excused. Late homework is not
Late Homework
accepted unless the lateness is excused for suitable reasons.
The University of Texas System and The University of Texas at Dallas
Student
have rules and regulations for the orderly and efficient conduct of
Conduct
university business. General information on student conduct and
and
discipline is contained in the UTD publication, A to Z Guide, provided to
Discipline
all registered students each academic year.
The faculty expects from students a high level of responsibility and
Academic
academic honesty. Students guilty of academic dishonesty are subject to
Integrity
disciplinary proceedings.
The University of Texas at Dallas encourages faculty to consider email
from students official only if it originates from a UTD student account.
This allows the university to maintain a high degree of confidence in the
Email identity of all individuals corresponding and in the security of the
transmitted information. UTD furnishes each student with a free email
account, and the Department of Information Resources at UTD provides a
method for students to forward their UTD email to other accounts.
Deadlines for withdrawal from courses are published in each semester's
course catalog. A faculty member cannot drop or withdraw a student.
Withdrawal
Rather, it is the student's responsibility to handle withdrawal procedures
from any class.
As per university policy, incomplete grades are granted only in the case of
work unavoidably missed (and excused) by the semester’s and not already
covered by the professor’s policy on missed work or activities, and only if
Incomplete
70% of the course work has been completed. An incomplete grade must
Grades
be resolved within eight weeks from the first day of the subsequent long
semester. Otherwise the incomplete grade becomes changed automatically
to the grade of F.

Disability Disability Services seeks to provide students with disabilities educational

Services opportunities equivalent to those of their non-disabled peers.
The University of Texas at Dallas excuses students from class or other
required activities for the purpose of travel to and observance of a
religious holy day for a religion whose places of worship are exempt from
Religious property tax under Section 11.20, Tax Code, Texas Code Annotated. In
Holy Days the case of such an absence, the student is encouraged to notify the
instructor as soon as possible, preferably in advance. Regarding missed
assignments, quizzes, tests, or exams, the student excused for such a
purpose will be covered by the professor’s policy for missed or late work.