ECON60052 · Cross-Section Econometrics

Cross-section econometrics

Free postgraduate study notes on cross-section and panel econometrics: the potential-outcomes framework for causal inference, robust and cluster-robust standard errors, difference-in-differences policy analysis, and the fixed- and random-effects panel estimators. Each lecture is a full worked study page with self-check questions.

These notes follow a single arc: from causality and treatment effects, through getting inference right under heteroskedasticity and clustering, to policy analysis with panel data and the advanced panel estimators. They are written for a postgraduate cross-section econometrics module (module code ECON60052) but suit anyone learning applied microeconometrics.

Lecture 1 Treatment effects & causal inference Potential outcomes, ATE and ATT, the selection-bias decomposition, randomisation, unconfoundedness and overlap, and matching estimators. Potential outcomesATE / ATTSelection biasMatching Lecture 2 Heteroskedasticity & clustering Why OLS standard errors fail, White/Eicker–Huber robust standard errors, testing, the intraclass correlation, the Moulton factor and cluster-robust inference. Robust SEsWhite testClusteringMoulton Lecture 3 Policy analysis with panel data Pooled cross sections, interaction terms, difference-in-differences, natural experiments and the common-trends assumption, plus two-period first differencing. Difference-in-differencesNatural experimentsFirst differencing Lecture 4 Advanced panel data methods First-difference, fixed-effects (within/LSDV) and random-effects estimators, the FE–FD comparison, the Hausman test and Mundlak's correlated random effects. Fixed effectsRandom effectsHausman testMundlak

A note on the first lecture: the treatment-effects material is provided as study notes only — there is no separate slide PDF for it. Lectures 2–4 include downloadable slides, linked on each page below and here.

Related resources on this site

These complement the cross-section material with the endogeneity and diagnostic-testing topics from the postgraduate econometric-methods lectures.

Diagnostic testing & robust inference — White and Breusch–Godfrey tests, HC and HAC/Newey–West standard errors (pairs with Lecture 2).
Endogeneity & instrumental variables (2SLS) — the standard remedy when a regressor is correlated with the error.
Weak & unidentified instruments — Stock–Yogo, Anderson–Rubin and LIML.
Econometrics explained — a plain-English map of the whole subject.
All teaching materials — time series, econometric methods and financial econometrics notes, slides and solutions.

Studying this for a module?

I tutor economics, econometrics and maths at university and postgraduate level — from weekly problem sets and past exams to dissertation methods. Book a free 30–45 minute consultation to talk through exactly where you are.

Book consultation Pricing & how it works