ECON400: pass the exams, not just read the notes
Your complete guide to University of Wisconsin-Madison's introduction to applied econometrics course. See where the marks are, work real practice questions, and study with an AI tutor that knows ECON400.
Sia generates ECON400 practice questions, walks through ols and inference step by step, and quizzes you on the material the exam weights most heavily.
Worked example
A regression of log(wage) on years of schooling gives a coefficient of 0.08 with standard error 0.01. Which statement is correct?
With a log outcome and a level regressor, a coefficient of 0.08 means approximately an 8% change in wage per year of schooling.
Option B misreads the log specification; option C confuses magnitude with significance.
Option D overreaches: without an experimental or quasi-experimental design, omitted ability bias means the estimate is an association, not a proven causal effect — the distinction the course's causal half is about.
The trap: Treating a precisely estimated coefficient as a causal effect. Precision says nothing about omitted variables. classic slip!
One exam decides 32% of your grade. Accommodation only for documented, pre-notified circumstances. This whole page is built around that.
Overview
What ECON400 is, and where it sits
ECON 400 Introduction to Applied Econometrics is UW-Madison's applied route into econometrics, taught in Fall 2026 by Christopher McKelvey with two teaching assistants. It follows ECON 310 and is the alternative to ECON 410: the department offers both, students take one, and the mathematics-emphasis and honors majors are required to take 410 instead.
The syllabus sets the emphasis plainly: develop the skills to read empirical papers and apply methods to real data in Stata. The twenty-four lectures run from OLS and its assumptions through multiple regression, non-linearities, interaction terms, measurement error and limited dependent variables, then spend the second half on causal designs — randomized trials, difference-in-differences, panel data, instrumental variables and regression discontinuity — with readings from Stock and Watson.
The Fall 2026 grade is 16% attendance recorded by TopHat, 20% weekly problem sets with a Stata log, a 32% in-class midterm on 22 October and a 32% cumulative final on 12 December. Grades are curved upward only: each student receives the better of a fixed percentage scale and a percentile scale.
Always treat your own course outline and the exam timetable as authoritative.
Difficulty & time commitment
Is ECON400 hard, and how much time does it take?
ECON400 is manageable if you keep a weekly rhythm and treat the back half as the main event. The pattern is consistent: it starts gently and steepens, and the heaviest assessment is the part that separates grades.
The difficulty curve and the assessment weighting point the same way: the back half is harder and worth more. Front-loading effort there is the highest-return decision in the course.
Is this course for you
Who tends to do well, and who tends to struggle
You will likely do well if
- You want to read empirical papers and run regressions rather than prove estimator properties.
- You will attend every class; 16% is simply for being there.
- You keep the weekly Stata problem sets moving; they are 20% and compound into exam readiness.
- You are comfortable with ECON 310 statistics.
You may struggle if
- You wanted the mathematical treatment; that is ECON 410, and you cannot take both.
- You spend the twelve excused absences early in the semester.
- You submit problem sets without the Stata log; full credit requires it.
- You treat the causal-design half as optional; it is half the final.
- For each method, write a one-line statement of the assumption that makes it causal and the situation that breaks it.
- Keep a Stata command sheet from the problem sets and bring it to revision.
- Practise reading regression tables from published papers, since a stated learning outcome is interpreting journal results.
- Work the midterm material to fluency before the causal designs start; they assume it.
Syllabus
The 11 topics, block by block
The exam-weight marker on each topic shows where the marks concentrate. The amber topics carry the highest exam weight.
T1 · OLS and goodness of fit
Lectures 2-3; S&W 4Ordinary least squares, R-squared, the regression assumptions and normality.
T2 · Inference and dummy variables
Lectures 4-5; S&W 5Hypothesis tests and confidence intervals for slopes; binary regressors; BLUE and heteroskedasticity.
T3 · Multiple regression
Lectures 6-8; S&W 6Adding controls; fit, properties and the nuances of interpretation.
T4 · Inference in multiple regression and model selection
Lectures 9-10; S&W 7Joint tests and choosing a specification.
T5 · Non-linearities and interactions
Lectures 11-12; S&W 8Logs, polynomials and interaction terms.
T6 · Data problems and limited dependent variables
Lectures 13-14; S&W 9, 11Measurement error, missing data, outliers; probit and logit.
T7 · Randomized control trials
Lectures 15-16; S&W 3.5Experiments as the causal benchmark.
T8 · Difference-in-differences
Lectures 17-18; S&W 10Before-after comparisons across treated and control groups.
T9 · Panel data
Lectures 19-20; S&W 10Fixed effects and within-unit variation.
T10 · Instrumental variables
Lectures 21-22; S&W 12Endogeneity and the IV solution.
T11 · Regression discontinuity
Lectures 23-24; S&W pp. 494-495Thresholds as natural experiments.
How it's assessed
Assessment structure
| Component | Weight | Format & timing |
|---|---|---|
| Attendance (TopHat, lectures and discussion) | 16% | TopHat check-in at every non-exam lecture and discussion section; twelve excused absences, further absences reduce the mark proportionally. Every class meeting. Location-verified attendance. |
| Weekly problem sets (Stata log required) | 20% | Weekly Stata problem sets submitted in Canvas with the Stata log attached. Weekly. 20% per day late deduction. |
| Midterm exam (in class, 22 October 2026) | 32% | In-class midterm during the regular lecture slot. 22 October 2026. Accommodation only for documented, pre-notified circumstances. |
| Final exam (cumulative, 12 December 2026, 2:45-4:45pm) | 32% | Cumulative final in the university exam block. 12 December 2026, 2:45-4:45pm. In person. |
- The four components sum to 100 and there is no separate hurdle. The curve is one-directional: the higher of the percentage scale (A from 92%) and the percentile scale (A for the top 20%) is awarded. Twelve absences are excused automatically; any further absence reduces the 16% attendance mark.
- One in-class midterm (22 October 2026) on the first half — regression, inference and specification — and a cumulative two-hour final (12 December 2026) that adds the causal-design half. Both are in person; accommodation requires documented, pre-notified circumstances.
This is an exam-cram course. With the exams at 64% of the grade and the midterm exam (in class, 22 october 2026) alone at 32%, your result is overwhelmingly decided by how well you perform under time pressure. Accommodation only for documented, pre-notified circumstances.
Final exam timing: During the examination period. Confirm the exact date and venue on your exam timetable.
How to actually pass it
A weekly rhythm, two checklists, and the traps to avoid
The course rewards consistency over cramming, and practice over re-reading. Here is the loop that works, then what to have nailed before each exam.
The weekly loop
Before the mid-semester checklist
- Interpret OLS coefficients, R-squared and standard errors.
- Run and read a t test and a joint F test.
- Explain heteroskedasticity and why robust standard errors are used.
- Interpret log, polynomial and interaction specifications.
Before the final heaviest topics
- State what makes an RCT the causal benchmark.
- Set up a difference-in-differences regression and its parallel-trends assumption.
- Explain fixed effects and what they remove.
- State the IV relevance and exclusion conditions and the RD identification idea.
The mistakes that cost marks
Correlation read as cause. A stated outcome is judging when a statistical association is causal; exams test the reasoning, not the command.
Missing Stata logs. Problem sets without the log do not receive full credit.
Burning excused absences. The syllabus warns that using the twelve early leaves no flexibility later.
Teaching team
Who teaches ECON400
The bios below are factual. We do not rate lecturers; any star ratings are submitted by students who have taken ECON400.
Teaching team as listed in the course materials reviewed. AskSia does not rate lecturers; star ratings are submitted by students who have taken ECON400.
Formula & concept sheet
The vocabulary and formulas you must own
- OLS
- Ordinary least squares: the line minimising squared residuals.
- R-squared
- The share of outcome variance explained by the regression.
- Heteroskedasticity
- Error variance that changes with the regressors; handled with robust standard errors.
- Omitted variable bias
- Bias from leaving out a variable correlated with both regressor and outcome.
- Interaction term
- A product of regressors letting one effect depend on another.
- Difference-in-differences
- The change in a treated group minus the change in a control group.
- Fixed effects
- Unit-specific intercepts absorbing time-invariant confounders.
- Instrumental variable
- A variable affecting the outcome only through the endogenous regressor.
- Regression discontinuity
- Identifying effects from a threshold rule on a running variable.
- Probit and logit
- Models for binary outcomes.
Set texts
The prescribed reading
The syllabus references map straight onto these.
Introduction to Econometrics
.
Where it fits
Prerequisites, related courses & why it matters
Prerequisite: ECON 310. Not open to students with credit for ECON 410. 4 credits; two 75-minute lectures and one 50-minute discussion per week.
Your ECON400 study toolkit
Study the course with Sia, not just read about it
Each tool already knows ECON400: your syllabus, your texts, and where the marks are. Grouped by how you study, from first contact to exam week.
FAQ
Frequently asked questions
Is ECON 400 hard?
Moderate on the six-factor rubric, and the easier of the two UW-Madison econometrics routes. Attendance and problem sets are 36% of the grade and the curve can only raise a grade.
What is the assessment breakdown?
Attendance 16%, weekly problem sets 20%, midterm 32% and final exam 32%, per the instructor's Fall 2026 syllabus.
ECON 400 or ECON 410?
They are alternatives, not a sequence. 400 emphasises applied technique and reading papers; 410 derives and proves results. Math-emphasis and honors majors must take 410.
Who teaches it?
Christopher McKelvey (Lecturer) in Fall 2026, with teaching assistants Julie Kim and Masahiro Nishida.
Do I need Stata?
Yes. Problem sets require a Stata log; UW-Madison's site licence makes Stata free from the Campus Software Library.
How is attendance taken?
Through TopHat with location verification at every non-exam class; students who decline location services sign in at the front at the end of class.
Study ECON400 with Sia
Work through ols, inference, multiple regression and the rest of the course with a tutor that knows it and quizzes you on the topics the assessments weight most heavily.
Start studying with Sia
