Econometrics Course

Lesson 1 • Probability and Distribution Review

Reviews probability rules, random variables, and key distributions used in econometrics. Connects distributional assumptions to later regression inference.

Lesson 2 • Statistical Inference Essentials

Covers estimation, hypothesis testing, and confidence intervals as tools for drawing conclusions from samples. Prepares students for regression-based inference.

Lesson 3 • Matrix Algebra for Econometrics

Introduces matrix operations essential for compact representation of regression models. Students gain fluency in matrix notation used throughout the course.

Lesson 4 • Role of Econometrics in Economics

Defines econometrics and distinguishes it from statistics and economic theory. Establishes why empirical testing of economic models matters for policy and research.

Lesson 1 • The Bivariate Regression Model

Specifies the population regression function and its stochastic component. Establishes the conceptual link between economic relationships and regression equations.

Lesson 2 • Goodness of Fit and Model Evaluation

Introduces R-squared, SST, SSE, and SSR decomposition to assess model fit. Students learn to distinguish statistical fit from economic meaningfulness.

Lesson 3 • Classical Assumptions of OLS

States the Gauss-Markov assumptions and explains their economic rationale. Connects assumption violations to biased or inefficient estimates.

Lesson 4 • Inference in Simple Regression

Derives sampling distributions of OLS estimators and constructs hypothesis tests and confidence intervals. Applies t-tests to slope and intercept coefficients.

Lesson 1 • OLS in Matrix Form

Derives the OLS estimator as the matrix formula and proves the Gauss-Markov theorem. Students apply matrix algebra to compute and verify estimates.

Lesson 2 • Multiple Regression Specification

Introduces the k-variable regression model and the role of ceteris paribus interpretation. Addresses how additional regressors control for confounding factors.

Lesson 3 • Hypothesis Testing in Multiple Regression

Extends t-tests to multiple coefficients and introduces the F-test for joint significance. Students test economic restrictions on model parameters.

Lesson 4 • Qualitative Regressors and Dummy Variables

Introduces binary and categorical variables as regressors and addresses the dummy variable trap. Students model group differences and structural shifts.

Lesson 1 • Autocorrelation in Regression Models

Addresses serially correlated errors common in time-series data and their effect on OLS efficiency. Introduces detection tests and corrective estimators.

Lesson 2 • Heteroskedasticity: Detection and Correction

Explains how non-constant error variance inflates standard errors and distorts inference. Covers formal tests and robust estimation strategies.

Lesson 3 • Multicollinearity: Diagnosis and Mitigation

Distinguishes perfect from near-perfect collinearity and its effect on coefficient precision. Provides practical strategies for reducing collinearity in applied work.

Lesson 4 • Measurement Error and Its Consequences

Analyzes how errors in variables bias OLS estimates and affect inference. Introduces instrumental variable intuition as a remedy.

Lesson 1 • Sources and Consequences of Endogeneity

Defines endogeneity and traces its three main sources: omitted variables, simultaneity, and measurement error. Demonstrates why OLS is inconsistent under endogeneity.

Lesson 2 • Applied IV in Economic Research

Examines classic IV applications in labor, health, and development economics. Students critically evaluate instrument validity in published studies.

Lesson 3 • Testing Instruments and Endogeneity

Covers the Hausman test for endogeneity and tests for instrument relevance and overidentification. Students validate instrument choices using formal diagnostics.

Lesson 4 • Instrumental Variables Estimation

Introduces the IV estimator, relevance and exogeneity conditions, and the Wald estimator for binary instruments. Students derive and compute IV estimates by hand.

Lesson 1 • Cointegration and Error Correction Models

Defines cointegration as a long-run equilibrium relationship and derives the error correction model. Students test for cointegration and estimate ECMs for economic data.

Lesson 2 • Unit Root Testing

Introduces the Dickey-Fuller and augmented Dickey-Fuller tests for unit roots in economic series. Students determine integration order and apply appropriate transformations.

Lesson 3 • Forecasting with Time-Series Models

Develops point and interval forecasts from ARIMA and ECM models and evaluates forecast accuracy. Students compare models using out-of-sample forecast metrics.

Lesson 4 • ARMA Models and Box-Jenkins Method

Covers AR, MA, and ARMA model specification, estimation, and diagnostic checking. Students apply the Box-Jenkins identification strategy to economic time series.

Lesson 1 • Structure and Advantages of Panel Data

Describes balanced and unbalanced panels and explains how repeated observations control for unobserved heterogeneity. Motivates panel methods over pooled OLS.

Lesson 2 • Random Effects and Hausman Test

Introduces the random effects estimator as a GLS approach and compares it to fixed effects. The Hausman test guides model selection between FE and RE.

Lesson 3 • Fixed Effects Estimation

Derives the within estimator by demeaning to eliminate time-invariant unobserved effects. Students implement FE models and interpret within-unit variation.

Lesson 4 • Dynamic Panel and GMM Estimation

Addresses dynamic panels with lagged dependent variables and the Nickell bias in FE models. Introduces Arellano-Bond GMM as the standard solution.

Lesson 1 • Censored and Truncated Regression

Addresses sample selection and censoring using the Tobit model and Heckman selection correction. Students identify when standard OLS produces biased estimates.

Lesson 2 • Count Data Models

Models non-negative integer outcomes using Poisson and negative binomial regression. Students test for overdispersion and select the appropriate count model.

Lesson 3 • Model Fit and Specification for Binary Models

Evaluates binary models using pseudo-R-squared, likelihood ratio tests, and classification metrics. Students diagnose misspecification in probit and logit models.

Lesson 4 • Ordered and Multinomial Models

Extends binary choice to ordered outcomes (ordered probit/logit) and unordered categories (multinomial logit). Students interpret threshold parameters and relative probabilities.