Instructor:	Dr. Brian Habing, Associate Professor office: 203 LeConte e-mail: habing*stat.sc.edu MSN messenger: brian@habing.com AOL or Yahoo messenger: DrStatpsy FAX: (803)777-4048 office phone: (803)777-3578 home phone: (803)739-2686 (9am to 10pm only) Office Hours: whenever the door is open, by appointment, and Monday/Wednesday/Friday 10:10-11:40 Website: http://www.stat.sc.edu/~habing/courses/518F07.html
Bulletin Description:	518—Nonparametric Statistical Methods (3) (Prereq: A grade of C or better in STAT 515 or equivalent) Application of nonparametric statistical methods rather than mathematical development. Levels of measurement, comparisons of two independent populations, comparisons of two dependent populations, test of fit, nonparametric analysis of variance, and correlation.
Note:	The deadline for dropping the course without failing is Thursday, October 4th. Thursday, October 11th and Thursday, November 22nd are holidays.
Purpose of Course:	To introduce the principles and applications of commonly used nonparametric methods. To compare these methods to their parametric counterparts. To introduce the basic methods for analyzing contingency tables.
Expectations:	All students are expected to: Attend class regularly, asking questions when clarification is needed and participating in any in-class activities. Read the pages covered in each class before the following class period. Attempt all of the assigned homework problems and turn them in at the start of the class in which they are due Use the resource of their fellow students and their instructor to seek answers to questions that arise in class, in the readings, and on the homework
Required Text:	Practical Nonparametric Statistics (3rd Edition), by W.J. Conover, Wiley, 1999. The readings associated with each class will be listed on the board at the beginning of class. After the class, it is the student's responsibility to read over the listed pages and to seek clarification on any difficulties found. (Either stop by during office hours, send an e-mail, or ask before or after the next class.)
Computers:	Use of a computer is required for the analysis of any sizable data set. This course will use the packages SAS and R. R is available for free (see the course page for downloading instructions). SAS is available in LeConte 124 and 303A, as well as in PSC 102. The university has discounted rates for SAS (see the link on the course page). NO PREVIOUS KNOWLEDGE OF R or SAS IS ASSUMED.
Course Outline:	Topics Covered Chapters Time         Probability and Statistical Inference – probability and counting rules, discrete random variables, continuous random variables, use of SAS and R, properties of estimators, properties of hypothesis tests 1-2 3 weeks Tests Based on the Binomial Distribution – the binomial and quantile test, the sign test, McNemar's test 3 1.5 weeks Methods Based on Ranks – Mann-Whitney test, Kruskal-Wallis test, squared rank test, measures of rank correlation, nonparametric linear regression, Wilcoxon signed ranks test, Friedman test 5 4 weeks Goodness of Fit Tests – Kolmogorov goodness of fit test, Kolmogorov test for two samples 6 1 week Categorical Data – chi-squared goodness of fit, chi-square test for r by c contingency tables, Mantel-Haenszel test, Cochran's test for related observations, measures of dependence, loglinear models 4 3.5 weeks
Grades:	The course grade is determined based on the performance on the quizzes (30%), homework (40%), and final project (30%). The grading scale may be adjusted as needed, but will not be made more difficult than: Letter Minimum Grade Percent A 90 B+ 87 B 80 C+ 77 C 70 D+ 67 D 60 F 0 There is no "extra credit". Any deviations from the above grading scheme will be to the benefit of the students.
Homework:	Each of the twelve homework assignments are due at the beginning of the class period it was assigned for. Late homework is not accepted. Homework will be assigned at least one week in advance in class, and will also be posted on the class website. The writing on the homework must be legible, the work used to obtain the answers must be shown and correct, and the final answers must be clearly indicated in order to receive full credit. Extra points may be deducted for violating any of the following: Write on one side of the paper only. Multiple pages must be stapled together. No clips. Copies of the computer code or menu options used must be included with any homework requiring R or SAS. Extraneous pages of SAS/R output should not be turned in. Your name should be on the first page of the assignment. You MAY consult with other students on the homework assignments (e.g. you can ask each other for advice and may work on the big picture together, but you should write up the details yourself). The lowest two homework grades will be discarded.
Quizzes:	Seven (approximately 15 minute) quizzes will be given during the semester. One will be given at the beginning of class on each of September 6th, September 20th, October 4th, October 18th, November 1st, November 24th, and November 29th. The lowest two quiz grades will be dropped.
Final Project:	An individual final project will be due by 5:30pm on Saturday, December 15th. The project will contain the analysis of a real data set of interest using both parametric and nonparametric methods, and a simulation study to compare the two procedures used. The outline of all required programs will be provided. The details on the project will be announced on October 2nd.
Graduate Credit:	Students taking the course for graduate credit will be assigned extra questions as part of the homework and will have additional requirements on the final project (thus making their grading stricter).
Complaints and Comments:	While there are end of semester evaluation forms, they come far too late to resolve any difficulties experienced in the class. All complaints should be raised by either speaking with me directly, or by anonymously leaving a message in my mailbox in 216 LeConte.