STAT 368
Modern Nonparametric Statistics Spring 2020
Division III Quantative/Formal Reasoning
This is not the current course catalog

Class Details

Many statistical procedures and tools are based on a set of assumptions, such as normality or other parametric models. But, what if some or all of these assumptions are not valid and the adopted models are miss-specified? This question leads to an active and fascinating field in modern statistics called nonparametric statistics, where few assumptions are made on data’s distribution or the model structure to ensure great model flexibility and robustness. In this course, we start with a brief overview of classic rank-based tests (Wilcoxon, K-S test), and focus primarily on modern nonparametric inferential techniques, such as nonparametric density estimation, nonparametric regression, selection of smoothing parameter (cross-validation), bootstrap, randomization-based inference, clustering, and nonparametric Bayes. Throughout the semester we will examine these new methodologies and apply them on simulated and real datasets using R.
The Class: Format: lecture
Limit: 30
Expected: 15
Class#: 3566
Grading: yes pass/fail option, yes fifth course option
Requirements/Evaluation: performance on exams, homework, and a project
Prerequisites: STAT 201 and STAT 346, or permission of instructor.
Enrollment Preferences: senior Statistics majors
Distributions: Division III Quantative/Formal Reasoning
QFR Notes: This is a statistics class with a focus on mathematical, computational, and data analysis skills as well as appropriate practical application of analysis methods.

Class Grid

Course Catalog Archive Search



Start Time
End Time