|
MATH 452/552 Statistical
Computing
Spring 2008
Call No. 07298/07299
Home | Mathematics Department | Ohio
University
Class homepage: http://www.math.ohiou.edu/~shen/spring2008/class452/class452.html
Time:
MTWTh
05:10-06:00 PM, 422 Morton Hall
Instructor: Dr. Annie X. Shen
Office: 571 Morton Hall
Office Hours: Tuesday,
Wednesday and Thursday, 04:00-5:00 pm or by appointments
Email: shen@math.ohiou.edu
Phone: 740-593-1288
Class
materials available at 
Catalog Description. Introduction to computational statistics; Monte
Carlo methods, bootstrap, data partitioning methods, EM algorithm,
probability density estimation, Markov Chain Monte Carlo methods.
• Prerequisite. MATH 450B/550B or equivalent.
The course aims to introduce the data analysis software, Statistics Toolbox
in MatLab and make computational statistics techniques available to engineers
and scientists. It emphasizes implementation rather than theory. Because
it is computational in nature, knowledge in matrix theory/linear algebra and
working experiences of at least one of programming languages such as C/C++,
FORTRAN or MatLab are required. A student who is not properly prepared for
this class generally performs poorly. Homework and exams may involve the
use of MATLAB and C/C++ programming at elementary level.
• Textbook
Computational Statistics Handbook with Matlab, Computational Statistics
Handbook with Matlab, 2nd, Wendy L. Martinez and Angel R. Martinez, CRC/Chapman
Hall, 12/20/2007.
• Software
1. Computational Statistics Toolbox for MATLAB.
The toolbox can be downloaded at http://www.pi-sigma.info/CS2E.htm
2. Statistics Toolbox in Matlab.
This software is available in public computer labs across campus,
and in the Morton Hall computer labs. You can find the menu at: http://www.mathworks.com/products/statistics/
.
• Coverage. Selected topics from:
Chapter 4 Generating random variables
Chapter 5 Exploratory data analysis
Chapter 7. Monte Carlo Methods for inferential statistics, bootstrap
methods
Chapter 9. Probability density estimation and EM algorithm
Chapter 14* Markov Chain Monte Carlo Methods.
• Grading Policy
All exams should be done independently; otherwise you will be guilty of
plagiarism. A student who is caught cheating will get an F for the course
and will be referred to student judiciaries for further action. Attendance
at the scheduled examinations and final is required except medical reasons.
If you have documented disability and need special accommodation, please
see me within 2 weeks of the quarter begins.
Letter grade for the quarter will be assigned
Score [ 0, 60) [60, 70) [70, 80) [80, 90) [90,100]
Letter grade F D C B
A
and based on: Homework 20%; Midterm (30%) Final Exam (50%)
• Assignments
Assignments are graded on accuracy, completeness, and good programming
methods. Document your programs with comments. Your work should be clear
enough that any reader could reproduce your results. Do not submit everything
that appears in the console. Instead, copy the essential output and paste
it into a report. It is convenient to use the function dairy and script editor
in Matlab to prepare the report, or use and windows word processor. For
each assignment submit
(1) Electronic copy of your program (s);
(2) Report containing neatly organized and annotated output; including
specific answers to any questions;
(3) Attach any graphs required, labeled as Figure 1, Figure 2, etc. and
reference by figure number in part (2) Each figure has to have a caption/title.
- Some interesting
websites
|