Syllabus

STAT 45203 · Resampling, Nonparametric, and Robust Methods · Fall 2026

This page is a public rendering of the course syllabus for convenient reading. Blackboard Ultra is authoritative for scheduling, assignment links, project checkpoints, official grades, and any change announced during the semester. Where this page and Blackboard differ, Blackboard governs.

Course at a glance

Course	STAT 45203 — Resampling, Nonparametric, and Robust Methods
Section	01 · In-person · Mon/Wed/Fri
Term	Fall 2026 · Aug 24 – Dec 15, 2026 (last class meeting Dec 7)
Instructor	Matthew Hester · `mhester@ualr.edu` (best contact)
Office	Math Assistance Center (MAC), Library Commons — Ottenheimer Library, 1st floor
Software	R · RStudio / Posit Cloud · Quarto (all free)

Course description

This course studies statistical methods that remain useful when standard parametric assumptions are questionable, hard to justify, or not the main point of the analysis. We begin from the question what can we responsibly infer with weaker assumptions? and study empirical distributions, order statistics, ranks, permutation logic, randomization tests, bootstrap distributions and confidence intervals, rank-based one-sample/paired/two-sample methods, categorical and ordinal outcomes, robust summaries and outliers, robust-regression ideas, and simulation studies that compare how methods behave.

The emphasis throughout is reasoning, computation, comparison, and interpretation — not memorizing a decision chart. You will compare parametric and nonparametric conclusions without automatically treating either as more “correct,” and communicate what a method does, what it assumes, what it protects against, and what it cannot prove.

Course goals

By the end of the course, a student who has done the work should be able to:

Explain why assumption-light methods are useful, and distinguish parametric, nonparametric, resampling-based, randomization-based, and robust approaches.
Use empirical distributions, ranks, order statistics, and quantiles to summarize data.
Explain and carry out permutation and randomization tests, and read a p-value obtained by simulation.
Use the bootstrap to approximate sampling variability, construct and interpret bootstrap confidence intervals, and explain what the bootstrap assumes and when it can fail.
Apply and interpret rank-based methods for one-sample, paired, and two-sample problems, and analyze ordinal/categorical outcomes with methods matched to the measurement scale.
Compare means, medians, trimmed means, and other robust summaries; identify outliers and influential points without treating every unusual value as an error; and explain the idea of robust regression.
Use simulation studies to compare method behavior (Type I error, power, bias, coverage) under different data-generating conditions.
Communicate an assumption-light analysis clearly — including what it cannot prove.

Required materials

Primary: instructor notes, examples, and method guides (this website + Blackboard). These are the main source for weekly definitions, computational examples, simulation templates, and method-comparison guides.
Supplementary open materials: selected sections from open statistics, inference, modeling, and data-science resources for bootstrap inference, permutation tests, simulation, and regression review. See Open readings & attribution. Because the course draws on resampling, classical nonparametrics, robust statistics, and simulation-based inference, no single textbook carries the whole course.
Software: R, RStudio or Posit Cloud, and Quarto — all free. See Software setup.
Calculator: any non-graphing scientific calculator (a TI-84 also works). Phones may not be used as calculators on quizzes or exams.
Blackboard Ultra: the official course home for announcements, assignment links, due dates, and project checkpoints.

Not used: Cengage, WebAssign, MyLab, or any paid homework platform.

Prerequisites

A prior introductory statistics course or comparable preparation. You should be comfortable with variables, data tables, graphical and numerical summaries, confidence intervals, hypothesis tests, p-values, and basic regression interpretation. Prior probability/inference is helpful for sampling-variability intuition but is reviewed as needed. Prior R experience is helpful but not required — computational examples and labs are scaffolded.

Weekly rhythm

Most non-exam weeks meet three times:

Monday — method concept + checkpoint. Introduce a resampling / nonparametric / robust idea, work examples, short checkpoint near the end of class.
Wednesday — computation or simulation day. Implement the method, run simulations, inspect output, visualize behavior, compare methods.
Friday — quiz / comparison day. A short quiz on recent material, then compare parametric vs. nonparametric conclusions or critique an analysis.

Holiday, exam, and review weeks use an adjusted rhythm shown on the Schedule.

Assessments and grading

Graded components (weights below) are collected in class and through Blackboard. The specific prompts, rubrics, point values, and due dates are posted in Blackboard, which is authoritative.

Category	Weight	Shape (structure only)
Method checkpoints	10%	Short in-class artifacts most Mon/Wed; lightly graded; lowest 4 dropped
Weekly quizzes	10%	Short Friday quizzes on recent material; lowest 2 dropped
Homework & method reports	25%	Roughly every two weeks; mixed hand-calculation, interpretation, simulation checks, written conclusions; lowest 1 dropped
Resampling & robustness labs	15%	Short computational activities connecting method logic to code
Midterm exam	15%	Fri, Oct 9, 2026, in class; case- and problem-based; first-half material
Applied robust-methods project	15%	Final-third project comparing ≥2 methods on a real/simulated problem
Final exam	10%	Cumulative; university final-exam window Dec 9–15, 2026

Grade scale: A 90–100 · B 80–89 · C 70–79 · D 60–69 · F < 60.

The weights and drop rules above come from the printed syllabus and are public. Individual graded-item content, keys, rubrics, and exact due dates are not published here — they live in Blackboard.

Late work and make-ups

Homework, method reports, labs: accepted up to one week late with a flat 20% penalty; after one week, no credit unless the absence was excused.
Checkpoints and quizzes: the lowest 4 checkpoints and lowest 2 quizzes are dropped automatically; missed ones ordinarily go into the drop pool with no separate make-up.
Midterm make-ups: only with a documented excused absence and advance instructor coordination. The final follows university policy.
Project checkpoints: part of the method-comparison process; late checkpoints limit feedback before the final submission.

AI usage policy (summary)

AI assistants (ChatGPT, Claude, Copilot, Gemini, …) may be used as study/workflow aids — concept explanations, practice questions, debugging simulation code, visualization suggestions. They may not produce work you submit as your own, complete homework/project solutions, fabricate simulation results, or invent interpretations, and are prohibited during quizzes and exams unless explicitly allowed.

Resampling and nonparametric methods are especially easy to get wrong — permuting the wrong thing, resampling rows when dependence should be preserved, treating a bootstrap interval as assumption-free, confusing ranks with raw values. On any graded written work, lab, code, or project component, include a brief AI Use Note:

Tool — which assistant, with approximate date/version.
Purpose — what you used it for.
Verification — how you checked, tested, revised, or validated the output.

Verification is the load-bearing line (rerun the code, check the resampling scheme, hand-work a small case, confirm what is held fixed under the null, compare to class notes).

Calendar & technology

Non-graphing scientific calculator recommended (TI-84 works); phones may not be used as calculators on quizzes/exams. During quizzes and exams, only explicitly allowed technology may be used.
R / Posit Cloud / Quarto / spreadsheets / browser tools may be used for labs, simulations, and the project. When software is used, the goal is to support statistical reasoning, not avoid it — you are expected to explain what is being resampled or ranked, what assumptions remain, and how the output supports the conclusion.

Key dates: Classes begin Aug 24 · Labor Day (no class) Sep 7 · Midterm Oct 9 · Last day to drop Oct 20 · Fall break Nov 22–28 · Last class Dec 7 · Final-exam window Dec 9–15. Full pacing on the Schedule.

Support & policies

Math Assistance Center (MAC): Library Commons — Ottenheimer Library, 1st floor · tutoring.ualr.edu
Office hours: best for method-choice questions, resampling logic, simulation debugging, interpretation, and project planning.
Writing support: recommended for making method-comparison reports clearer and more audience-aware.
University policies: Disability Resource Center · Academic Integrity · University Email · Inclement Weather. This syllabus may be updated for clarity or pacing; changes are announced via UA Little Rock email and posted in Blackboard.