Tessella Press Sign in
Econometrics and Empirical Analysis cover

Econometrics and Empirical Analysis

Mark Humphery-Jenner

Read each chapter, then work through auto-marked multiple-choice and numerical questions — plus written answers with instant feedback against the author's model solutions.

Read Chapter 1 Practice Chapter 1 Unlock all chapters Print & Kindle ↗

As an Amazon Associate, Tessella Press earns from qualifying purchases.

About this book

Regression is the workhorse of empirical research — and the easiest tool in the kit to use badly. Econometrics and Empirical Analysis is a practical course in doing regression analysis properly: what the models assume, how to tell when those assumptions fail, and how to move, carefully, from correlation to causation.

The book opens with OLS regression from first principles — the assumptions, the derivation of the estimator, hypothesis testing, and how to read real regression output — before turning to the things that go wrong in practice: heteroscedasticity, multicollinearity, omitted variables, outliers, and how to detect and fix each one. From there it covers binary and discrete choice models (logit, probit, multinomial and ordered outcomes), panel data (fixed effects, random effects, and Fama-MacBeth), and the modern causal-inference toolkit: instrumental variables and 2SLS, difference-in-difference, and regression discontinuity designs. Two applied chapters put the methods to work where empirical finance actually uses them — estimating CAPM betas and factor models, and running event studies. Every technique is developed through worked examples with real regression output, not abstraction.

What the online companion includes

Written for university courses in econometrics and empirical methods, for research students facing their first serious regressions, and for practitioners in finance and economics who need to read — and challenge — empirical results. Also available in print and Kindle editions.

1 OLS Regression Basics 50 questions · free sample
Open
2 OLS Regression Diagnostics and issues 50 questions
🔒
3 Binary response, and discrete choice variables 50 questions
🔒
4 Panel Data 50 questions
🔒
5 Causal inference and identification: From correlation to causation 50 questions
🔒
6 Difference-in-Difference and Regression Discontinuity Designs 50 questions
🔒
7 Example: CAPM, Single index models, Factor models 40 questions
🔒
8 Example: Event Studies 40 questions
🔒