Schedule (topic flow)

A 15-week pacing plan — Blackboard is authoritative for dates and deadlines

Important

This is a pacing plan, not a calendar. It shows the order in which the methods unfold and roughly when, so you can read ahead and see where you are headed. It is not the authoritative calendar: exact dates, deadlines, and any mid-term adjustments live in Blackboard (the LMS), which is authoritative. A few fixed semester dates are noted for orientation only — confirm everything against Blackboard.

The course runs fifteen weeks in four parts. Each part below lists its weeks, the theme for each week, and the throughline — the question that carries the week. The theme links to that week’s note.

Part I — Foundations of assumption-light reasoning (Weeks 1–2)

We set up the frame: why weaker assumptions are useful, and the empirical objects — order statistics, ranks, and the ECDF — that the rest of the course is built on.

Week	Theme	The throughline
1	Why assumption-light methods?	What is fragile in a standard analysis, and what do we do when the model is in doubt?
2	Order statistics, ranks & empirical distributions	How do quantiles, ranks, and the ECDF summarize data without a model?

Part II — Resampling inference (Weeks 3–6)

We build inference from the data itself: shuffle labels to test, resample to estimate uncertainty.

Week	Theme	The throughline
3	Permutation logic	Under the null, what stays fixed and what is exchangeable — and what does shuffling test?
4	Randomization tests	When the reshuffle mimics the assignment mechanism, what does the test license?
5	Bootstrap distributions	How does resampling with replacement approximate a statistic’s sampling variability?
6	Bootstrap confidence intervals	How do we turn a bootstrap distribution into an interval — and when does it fail?

Note: Labor Day (Mon, Sep 7) falls in Week 3 — no class that Monday, so the week runs Wednesday and Friday.

Part III — Rank-based & ordinal methods (Weeks 7–9)

We move to ranks: one-sample and paired, two-sample, and ordered categories — methods that use order, not the raw numbers.

Week	Theme	The throughline
7	Rank-based one-sample & paired methods (midterm)	How do the sign test and the signed-rank test trade assumptions for one-sample/paired data?
8	Two-sample rank methods	What does the rank-sum test claim, and why is it a shift, not a difference in means?
9	Categorical & ordinal outcomes	How do we respect the measurement scale of an ordered categorical outcome?

Note: the midterm is Friday, Oct 9 (in class), during Week 7. It covers empirical distributions through early rank-based methods. No graded content appears on this public site — see Blackboard.

Part IV — Robustness, comparison & synthesis (Weeks 10–15)

We add robustness — resistant summaries and robust regression — then compare methods, simulate their behavior, and synthesize.

Week	Theme	The throughline
10	Robust summaries & outliers	Which summaries resist contamination, and how do we find outliers without auto-deleting them?
11	Robust regression ideas	Why does least squares break under contamination, and how do robust fits hold?
12	Comparing parametric & nonparametric conclusions	When do the methods agree, when do they diverge, and how do we report the difference?
13	Simulation study of method behavior	How do Type I error, power, and coverage compare across data-generating processes?
14	Applied robust-methods report workshop	How do you compare methods on a real problem and write an honestly-bounded report?
15	Final review	How do resampling, ranks, robustness, and simulation fit as one assumption-light practice?

Note: fall break is Nov 22–28 (no classes), between Weeks 13 and 14. The last class is Mon, Dec 7, the consultation day is Dec 8, and the final-exam window is Dec 9–15 (exact block via Blackboard).

Note

Again: this page is a pacing plan to help you read ahead. For the real calendar — every date, every deadline, and any adjustment during the term — Blackboard (the LMS) is authoritative.