STAT 320: Methods of Non-Parametric Statistics
Prerequisite: STAT 200 or 301
Credit Hours: (3)
Examines distribution free analogs of many classical statistical tests. Topics include tests based on binomial distribution; tests based on Fisher's method of randomization; goodness of fit tests; two sample tests and correlation procedures. Use of modern computer software to analyze real world data.
Detailed Description of Content of the Course
The purpose of this course is to provide the student with an understanding of the techniques that can be used to analyze nonnormal data. The procedures given are especially useful with small samples. Topics include sign test, Wilcoxn rank sum and signed rank tests for one and two samples, Mann-Whitney test, Friedman test, Kruskal-Wallis test, Spearman correlation, contingency table analysis, tests for runs and randomness. The course is appropriate for statistics majors as well as others interested in data analysis.
Detailed Description of Conduct of the Course
The lecture format is used in the course. When possible in-class experiments are conducted to obtain real life data for use in illustrating the concepts presented. Homework assignments are used primarily for practice purposes.
Goals and Objectives of the Course
The objectives of the course are to provide alternatives to the usual normal theory statistics. This should equip the student to handle in an appropriate manner most data sets that they might encounter.
Tests are a mix between closed book in-class questions and take-home problems. The final exam is comprehensive.
Other Course Information
The course is offered primarily for mathematics students with a concentration in statistics. However, it is of interest to any student whose research will involve statistical methods.
Review and Approval
Sept. 2001 Review Stephen Corwin, Chair