Data Analysis -- Math 736
Fall 2007
"The manipulation of statistical formulas is no substitute for knowing what one is doing."
~Hubert M. Blalock, Jr.
Instructor: Michelle Lacey
Office: Gibson 417C
Phone: (504) 862-3439
E-mail: mlacey@math.tulane.edu
Office Hours: TBD
Course Website: http://math.tulane.edu/~mlacey/DataAnalysis
Prerequisites
Math 301/601 (or equivalent background in Probability and Statistics) and Math 604/304 (or equivalent background in Linear Models) must be completed prior to enrollment in Math 736. Math 602/724 is also recommended but may be taken concurrently. Exceptions to these prerequisites may be granted by permission of the instructor.
Monday and Wednesday 3:00-4:15 in Gibson 325.
Required text: W.N. Venables and B.D. Ripley, Modern Applied Statistics with S, Fourth Edition (Springer, 2002).
Interactive handouts will also be regularly distributed in lecture and posted on the course website.
Software: Students with personal computers should install the R software package. This free package is available for Windows, Mac, and Linux.
For Windows users (98 or higher):
From the web page http://cran.cnr.berkeley.edu/bin/windows/base/, download the file
"R-2.5.1-win32.exe". Launching this file will start a standard setup program.
For Mac and Linux users:
Follow the links from the http://cran.us.r-project.org/ for the appropriate package.
Additional materials: The R website: http://www.r-project.org/
The "Documentation" area of this site contains a wealth of information, including manuals, FAQs, and links to additional publications.
Course Description
This course provides an applied approach to the statistical analysis of datasets using the R software package. The R environment, which is an Open Source system based on the S language, is one of the most versatile and powerful tools available for statistical data analysis, and is widely used in both academic and industrial research.
While statistical methods are based on theoretical concepts, it is important to remember that the primary purpose of the field of statistics is to better understand observed data. There is a strong interdependence between statistics and applied research in every area: researchers conduct experiments according to statistical principles, and new statistical methods are developed in response to advances in technology and experimental practice. By the very nature of this relationship, it is clear that any course in the statistical analysis of data cannot be comprehensive. There are simply too many topics to cover them all in a single term, and it is likely that new approaches will be introduced at any given time. However, despite constant enhancements, the basic principle of data analysis remains the same: all methods are founded on statistical theory, and we assume that observed data can, in some way, be understood by an underlying stochastic model.
In this spirit, the objective of the course is to provide a foundation in data analysis, providing exposure to a variety of methods. Theory will be introduced as needed, but our primary focus will be the application of methods to actual datasets (and the interpretation of the results). Facility with data manipulation and knowledge of the R language will also be important by-products of the course, since nearly every dataset requires some degree of "preprocessing" before one can employ analytical tools.
Sequence of
Topics
Introduction to the R software package and programming language
Multiple regression and generalized linear models
Classification methods
Data visualization and clustering
Time series analysis
Spatial statistics
Regular attendance at lectures is expected. Due to the applied nature of this course, a large component of class time will be spent working in groups, and excessive absenteeism will be highly disruptive to the natural dynamic of the course.
Course grades will be based on homework assignments and
class participation. In lieu of a final exam, an independent
project will be
assigned.
This course operates in accordance with the Honor Code of
the College of the Liberal Arts and Sciences.
Any student requiring accommodations for a physical,
psychological, or learning disability must obtain proper documentation
through
the Office of Disability Services (ODS) at the Educational Resources
and
Counseling Center (ERC). Additional information is available at http://erc.tulane.edu.