Introductory Econometrics: A Modern Approach

Citation

Wooldridge, J. M. (2006). Introductory Econometrics: A Modern Approach (3rd ed.). Thomson/South-Western.

Chapter Summary

The third edition (2006), described in library records as 890 pages, has 19 chapters in three parts: regression analysis with cross-section data (Chapters 2 to 9), regression analysis with time series data (Chapters 10 to 12) and advanced topics (Chapters 13 to 19), with an opening chapter on the nature of econometrics and economic data. The publisher's synopsis describes the approach as bridging the gap between the mechanics of econometrics and the modern applications of econometrics, using a systematic approach motivated by the major problems facing applied researchers, with an emphasis on examples that give concrete reality to economic relationships and allow treatment of policy questions in an accessible framework.

Chapter 1: The Nature of Econometrics and Economic Data

The opening chapter introduces what econometrics is and the kinds of economic data that later chapters analyse. It sets up the vocabulary used for the three parts.

Part I: Regression Analysis with Cross-Section Data

Chapter 2: The Simple Regression Model. The starting point: regression with a single explanatory variable.

Chapter 3: Multiple Regression Analysis: Estimation. Estimation extended to several explanatory variables.

Chapter 4: Multiple Regression Analysis: Inference. Statistical inference for multiple regression.

Chapter 5: Multiple Regression Analysis: OLS Asymptotics. The large-sample properties of ordinary least squares.

Chapter 6: Multiple Regression Analysis: Further Issues. Additional practical issues in using multiple regression.

Chapter 7: Multiple Regression Analysis with Qualitative Information: Binary (or Dummy) Variables. Including qualitative information in regressions through dummy variables.

Chapter 8: Heteroskedasticity. What happens when error variances are not constant, and how to respond.

Chapter 9: More on Specification and Data Problems. Specification choices and problems in the data.

Part II: Regression Analysis with Time Series Data

Chapter 10: Basic Regression Analysis with Time Series Data. Regression when observations are ordered in time.

Chapter 11: Further Issues in Using OLS with Time Series Data. Complications specific to time series.

Chapter 12: Serial Correlation and Heteroskedasticity in Time Series Regressions. Errors that are correlated over time or have changing variance in a time series setting.

Part III: Advanced Topics

Chapter 13: Pooling Cross Sections across Time: Simple Panel Data Methods. Combining cross-sections from different periods and the first panel-data methods.

Chapter 14: Advanced Panel Data Methods. A more developed treatment of panel data.

Chapter 15: Instrumental Variables Estimation and Two Stage Least Squares. Estimation when an explanatory variable is correlated with the error term.

Chapter 16: Simultaneous Equations Models. An introduction to systems in which variables are determined jointly.

Chapter 17: Limited Dependent Variable Models and Sample Selection Corrections. Models for outcomes with restricted range, and corrections for non-random samples.

Chapter 18: Advanced Time Series Topics. Further topics in time series analysis.

Chapter 19: Carrying Out an Empirical Project. The closing chapter, on carrying out an empirical research project of one's own.

Key Concepts

Chapters 1–6

• Simple and multiple regression: estimating relationships between an outcome and one or several explanatory variables.
• Inference: testing hypotheses and building confidence statements about regression coefficients.
• OLS asymptotics: properties of ordinary least squares as the sample grows.

Chapters 7–9

• Binary (dummy) variables: representing qualitative information in a regression.
• Heteroskedasticity: non-constant error variance.
• Specification and data problems: the practical issues named in Chapter 9.

Chapters 10–12

• Time series regression: regression with data ordered over time.
• Serial correlation: correlation of errors across time.

Chapters 13–19

• Pooled cross-sections and panel data: data that follow units or combine samples over time.
• Instrumental variables and two stage least squares: methods for correlated regressors and errors.
• Simultaneous equations models: jointly determined variables.
• Limited dependent variables and sample selection: models for restricted outcomes and non-random samples.
• Empirical project: the practical skills that tie the methods to research.

Critical Analysis

Strengths

• Organisation by data type. Separating cross-section regression, time series regression and advanced topics gives a clear map, so a reader can see where each method belongs.
• Examples-first design. The publisher's synopsis states that the book emphasises examples that give concrete reality to economic relationships and let it treat interesting policy questions in an accessible framework.
• A place for advanced methods. Panel data, instrumental variables, simultaneous equations, limited dependent variables and advanced time series all have their own chapters in Part III.
• A chapter on carrying out a project. The closing chapter is devoted to carrying out an empirical project, so the book ends by turning the methods toward students' own research.

Limitations

• Edition. The third edition was published in 2006, and later editions of the book exist; students should check which edition their course uses, since chapter content and numbering may differ.
• Length. At about 890 pages the book covers more than one term. Part III is a menu from which a course picks topics.
• Treatment of causal designs. In this edition the contents place instrumental variables in Chapter 15 and panel methods in Chapters 13 and 14, with no dedicated chapters on difference-in-differences or regression discontinuity; the newer causal-inference designs are better learned from the more recent books in the course.

Where It Fits in the Course

In the econometrics list, Wooldridge is a full-length textbook that treats regression thoroughly across cross-section, time series and panel data. Studenmund's Using Econometrics is a gentler, intuition-led introduction to single-equation regression, and Newbold and colleagues' Statistics for Business and Economics covers the statistical foundations beforehand. Stock and Watson's Introduction to Econometrics is the other standard undergraduate textbook, built around empirical questions and real data sets, and can be used as a comparison or as a second source on topics where a different explanation helps.

Angrist and Pischke's Mastering 'Metrics is an accessible introduction to the methods economists use to estimate causal effects, and their Mostly Harmless Econometrics is a compact guide to regression, instrumental variables, differences-in-differences and regression discontinuity. Both are good companions to this book, which supplies the systematic regression foundation they build on.

How to Use This Book

The book is for undergraduates and master's students in economics and the social sciences who want a systematic course in regression. An introductory statistics course and basic calculus are a sensible starting point.

A practical plan:

• Core: Chapters 1 to 4 and 6 to 8, which give the cross-section regression toolkit. Chapter 5 on asymptotics can be read lightly on a first pass.
• Second layer: Chapter 9, then Chapters 10 to 12 for time series.
• Choose from Part III: Chapters 13 to 15 for panel data and instrumental variables are the most widely useful; Chapters 16 to 18 are for particular needs.
• Finish with Chapter 19 and an empirical project of your own, using statistical software on a real data set.
• Alongside: a current statistical package and, for causal designs beyond instrumental variables, one of the shorter applied books above.