01:220:322 Econometrics
The Complete Exam Bible · Spring 202601:220:322 Overview
- 3 credits
- Rutgers University
- Spring 2026
- Economics
- Stata-based problem sets
- Closed-book midterm and final
What the course is for
Econometrics at Rutgers University is taught as the bridge from studying economics to practising it. The syllabus describes a set of research tools for estimating and testing economic relationships, used across accounting, finance, marketing, management and the social sciences.
- Assessed by Homework, in-class quizzes, a midterm and a cumulative final, each weighted in the syllabus.
- Hardest step Interpreting a slope with the right functional form and naming what is held fixed.
- Most repeated question Signing omitted variable bias from two signs and checking it against a printout.
- How to prepare Rebuild each worked example by hand, then read real Stata output aloud as sentences.
How 01:220:322 is assessed
| Component | Weight | Format |
|---|---|---|
| Homework | 25% | Problem sets graded on completion and correctness |
| In-class quizzes | 5% | Five in-class quizzes graded on completion |
| Midterm | 35% | Closed-book exam; two double-sided note sheets and a calculator |
| Final | 35% | Closed-book cumulative exam; four note sheets and a calculator |
Weights are from the Spring 2026 syllabus and total 100%. Homework questions earn 90% of their points for completion, 95% if mostly correct and 100% if perfect, and late work loses points on a sliding scale. The final focuses on material after the midterm but is cumulative. There is no extra credit, and Economics majors must pass the course with a C or higher.
Current 01:220:322 dates
| Date | Item | Control |
|---|---|---|
| March 12, 2026 | Midterm | Spring 2026 offering, in the regular class time and room |
| May 7, 2026 | Final exam | Spring 2026 offering, 9:00 to 11:00am |
Dates are as published in the Spring 2026 syllabus. Confirm exact deadlines and submission settings in the live LMS.
What 01:220:322 covers
Ten chapters follow the lecture sequence from causal questions and a statistics review through simple and multiple regression, robust inference, functional form and dummy variables to instrumental variables and difference-in-differences.
Econometric Questions, Data and Causality
Ceteris paribus questions, potential outcomes, experiments versus observational data and the four data structuresWhat econometrics adds to statistics
Econometrics applies statistical methods to economic data so that models suggested by theory can be estimated and tested.
- Econometrics as the bridge from theory to data
- Causality versus correlation
- Potential outcomes and the missing counterfactual
- Experiments and observational data
- Cross sections, time series, pooled cross sections and panels
Probability and Statistics Review
Expected values, variances, covariance, the sampling distribution of the mean, tests, p-values and intervalsWhy a statistics review comes first
Regression inference is ordinary statistical inference applied to a slope.
- Expected value, variance and covariance
- Variance of a sum
- Unbiasedness and consistency
- Central limit theorem
- One-sided and two-sided tests
- Confidence intervals and p-values
Simple Linear Regression
The OLS line, residuals, R-squared, units and logarithms, and why OLS is unbiased under SLR assumptionsThe model and its error term
Simple regression writes the outcome as an intercept plus a slope times one explanatory variable plus an error. The error collects every other influence on the outcome.
- Population regression function and the error term
- OLS slope and intercept
- Residuals and their algebraic properties
- SST, SSE, SSR and R-squared
- Level and log interpretations
- SLR assumptions, unbiasedness and standard errors
Multiple Regression Estimation and Omitted Variables
Partial effects, MLR assumptions, perfect collinearity, omitted variable bias and the variance of a slopeWhy add regressors
Multiple regression holds observed factors fixed explicitly. Each slope becomes a partial effect: the change in the predicted outcome when that regressor changes and the others stay put.
- Partial effects and ceteris paribus interpretation
- MLR.1 to MLR.4 and unbiasedness
- Perfect collinearity versus multicollinearity
- Omitted variable bias and its sign
- Variance of a slope and its three ingredients
- Gauss-Markov and BLUE
Heteroskedasticity and Robust Standard Errors
What unequal error variance breaks, what it leaves unbiased, and robust inference in StataWhat heteroskedasticity is
Homoskedasticity, assumption MLR.5, says the variance of the error is the same whatever the values of the regressors.
- MLR.5 homoskedasticity
- Consequences for bias, efficiency and standard errors
- Heteroskedasticity-robust standard errors
- Robust t tests and intervals in Stata
Inference with t Tests, F Tests and Intervals
t statistics, critical values, p-values, confidence intervals, lincom and joint F tests of exclusion restrictionsOne coefficient at a time
The t statistic is the estimate minus the hypothesised value divided by the standard error.
- Critical values for t statistics
- One-sided and two-sided alternatives
- The p-value
- Confidence intervals and their interpretation
- Testing a difference with lincom
- Joint exclusion restrictions and the F test
Functional Form: Logs, Quadratics and Interactions
Log interpretations, marginal effects and turning points of quadratics, and interaction termsLogs revisited
Logarithms appear throughout the course. A log outcome makes a slope an approximate percentage change; a log regressor means the slope over one hundred is the effect of a one percent change; logs on both sides give an elasticity.
- Log and elasticity interpretations
- Quadratic marginal effects
- Turning points
- Testing linearity versus curvature
- Interaction terms and effects at the mean
- Centring
Dummy Variables and Qualitative Information
Group gaps, base groups, the dummy variable trap, dummy interactions and the linear probability modelCoding categories
A dummy variable equals one for a category and zero otherwise, and it is named after the category coded one.
- Dummy coefficients as group differences
- Base groups and the dummy variable trap
- Percentage gaps with log outcomes
- Several categories and ordinal variables
- Dummy interactions and cell means
- Linear probability model and its limits
Instrumental Variables and Two-Stage Least Squares
Endogeneity, relevance and exogeneity, the first stage, 2SLS in Stata and weak instrumentsThe endogeneity problem
An explanatory variable correlated with the error is endogenous, and OLS is then biased and inconsistent.
- Endogenous and exogenous regressors
- Instrument relevance and exogeneity
- Simple IV estimator
- First stage with controls
- Two-stage least squares in Stata
- Weak instruments and the first-stage F
Pooled Cross Sections and Difference-in-Differences
Year dummies in pooled samples, the two-by-two DID table, the regression form and parallel trendsPooled cross sections
Independent random samples drawn in different periods can be stacked into a pooled cross section.
- Pooled cross sections
- Year dummies and base years
- Interacting regressors with time
- The two-by-two difference-in-differences table
- Regression form of DID
- Parallel trends
By the end, students should estimate and interpret linear regressions in a statistical package, distinguish economic from statistical importance, and critique regression results in applied papers for a reader who is not trained as an economist.
How the material is organised
The lecture sequence starts with causal questions and data structures, reviews probability and statistics, and then builds regression in layers: one regressor, several regressors with omitted variable bias, heteroskedasticity, and inference with t tests, F tests and intervals.
The second half extends the model with logarithms, quadratics and interactions, dummy variables and the linear probability model, and closes with two causal designs, instrumental variables and difference-in-differences.
The textbook is Wooldridge's Introductory Econometrics, and Stata is used in lectures and problem sets.
What the exams look like
Both exams are closed book with a calculator and a limited number of note sheets.
They open with short multiple-choice items and then spend most of their points on a regression printout: writing the population model and the fitted equation, interpreting slopes with the right units, judging the zero conditional mean assumption, testing hypotheses and explaining how an estimate changes when a variable is added.
The midterm also tests a hypothesis about a population mean.
Where students lose points
The recurring slips are predictable: forgetting to divide a level-log slope by one hundred, calling heteroskedastic estimates biased, treating a high correlation as perfect collinearity, writing hypotheses about estimates instead of parameters, testing a variable that enters with a square or an interaction using a single t statistic, and confusing parallel trends with equal starting levels.
Each chapter of this guide flags the slip where it arises.
How to use this guide
Work one chapter at a time. Read the method, redo the worked example with a calculator, then attempt the practice cards before reading the answers. Build your permitted note sheets from the formula boxes as you go, and finish with the mixed practice set and the exam-morning checklist.
Worked example · free
Interpreting a regression printout
- 3Density (level-log): holding car ownership fixed, 1% higher density is associated with a transit share higher by 9.20/100 = 0.092 percentage points on average.
- 2Car ownership: holding density fixed, one more car per household is associated with a transit share lower by 6.50 percentage points on average.
- 3Dropping car ownership: its coefficient is negative and it is negatively correlated with density, so the short-regression density slope is biased upward and would exceed 9.20.
Key terms
- Ordinary Least Squares
- The method that chooses coefficients to minimise the sum of squared residuals.
- Zero Conditional Mean
- The assumption that the error has mean zero whatever the values of the regressors.
- Omitted Variable Bias
- Bias in a slope caused by leaving out a relevant variable correlated with an included regressor.
- Heteroskedasticity
- Error variance that changes with the regressors, which invalidates the usual standard errors.
- Standard Error
- The estimated standard deviation of an estimator, used to build tests and intervals.
- Joint F Test
- A test that several coefficients equal zero at the same time.
- Dummy Variable
- A zero-one variable that marks membership of a category.
- Instrumental Variable
- A variable that shifts an endogenous regressor without being related to the error.
- Parallel Trends
- The assumption that treated and control groups would have changed alike without treatment.
01:220:322 FAQ
How is the final grade in Econometrics at Rutgers weighted?
Homework counts for 25%, five in-class quizzes for 5%, the midterm for 35% and the final for 35%, according to the Spring 2026 syllabus. Quizzes are graded on completion and homework on completion and correctness.
What can I bring to the exams?
Both exams are closed book. The syllabus allows two sheets of notes written on both sides and a calculator for the midterm, and four sheets of notes and a calculator for the final exam in May.
Is the final exam cumulative?
Yes. The syllabus says the final focuses on material after the midterm, but because later topics build on regression basics, it is cumulative in practice and earlier methods reappear.
How is homework scored?
Each question has equal weight. Completing it earns 90% of its points, a mostly correct answer with some mistakes earns 95%, and a perfect answer earns 100%, so attempting everything guarantees at least 90%.
What software does the course use?
Stata. It is introduced in lectures and problem sets without assuming prior experience, and it is available through the university virtual computer labs as well as by student license.
Which topics does the course cover after the midterm?
The lecture sequence continues with t and F inference, logarithms, quadratics and interactions, dummy variables and the linear probability model, and then two causal designs: instrumental variables and difference-in-differences.
What background is expected before taking the course?
Principles of microeconomics and macroeconomics, Calculus I and an introductory statistics course. The syllabus strongly suggests reviewing that material at the beginning of the semester, before regression starts.
Questions students ask in each 01:220:322 chapter
- Why is correlation not enough in economics?
- People and firms choose their treatments, so the groups being compared usually differ in other ways. Those hidden differences move outcomes too, and the raw association cannot separate them from the effect. Chapter 01 · Econometric Questions, Data and Causality
- When do I use 1.645 rather than 1.96?
- Use 1.645 for a one-sided alternative at 5%, where all the rejection probability sits in one tail. A two-sided alternative splits the 5% across two tails, which pushes the cutoff out to 1.96. Chapter 02 · Probability and Statistics Review
- Does a low R-squared mean the regression is wrong?
- No. It means many factors other than x move the outcome, which is typical for data on individuals. The slope can still be unbiased; bias depends on whether the omitted factors are correlated with x. Chapter 03 · Simple Linear Regression
- Is a correlation of 0.9 between two regressors a violation?
- No. The no perfect collinearity assumption rules out only exact linear relations. A correlation of 0.9 leaves the estimates unbiased but makes them less precise, which shows up as larger standard errors. Chapter 04 · Multiple Regression Estimation and Omitted Variables
- Does heteroskedasticity make OLS biased?
- No. Unbiasedness relies on MLR.1 to MLR.4 only. What fails is the usual variance formula, so the reported standard errors and the tests built on them cannot be trusted. Chapter 05 · Heteroskedasticity and Robust Standard Errors
- Why can I not test two coefficients with two separate t tests?
- Each t test holds the other coefficient unrestricted, so together they do not test the joint claim, and correlated regressors can make both look insignificant. The F test evaluates the restrictions at once. Chapter 06 · Inference with t Tests, F Tests and Intervals
- Why is the coefficient on x alone not the effect of x in a quadratic?
- With a squared term the slope changes with x. The linear coefficient is the slope only at x equal to zero, so the effect must be computed at a value that matters, such as the mean. Chapter 07 · Functional Form: Logs, Quadratics and Interactions
- Why can I not include both a male and a female dummy?
- With an intercept in the model the two dummies add up to one for everyone, an exact linear relation with the constant. That is perfect collinearity, and software will drop one of them. Chapter 08 · Dummy Variables and Qualitative Information
- Which instrument condition can the data test?
- Relevance. Regress the endogenous variable on the instrument and all exogenous controls and test the instrument's coefficient. Exogeneity concerns the unobserved error and must be defended with an argument. Chapter 09 · Instrumental Variables and Two-Stage Least Squares
- Do the treated and control groups need the same starting level?
- No. Different starting levels are absorbed by the treated-group dummy. What matters is that, without the treatment, both groups would have moved by the same amount over time. Chapter 10 · Pooled Cross Sections and Difference-in-Differences
How to study for the exam
Study in two passes that match the two exams. For each chapter, redo the worked example by hand, write every slope interpretation as a full sentence with units and what is held fixed, and add the formula box to your note sheet. Before each exam, practise reading an unfamiliar regression printout aloud and signing the bias from an omitted variable.
Your AI Economics tutor for 01:220:322
Stuck on a hard 01:220:322 question? Sia is AskSia’s AI Economics tutor — ask any 01:220:322 Econometrics question and get a clear, step-by-step explanation grounded in how the course is actually taught and assessed. Read this whole study guide free, then take your hardest questions to Sia.
