Econometric Theory and Methods provides a unified treatment of modern econometric theory and practical econometric methods. The geometrical approach to least squares is emphasized, as is the method of moments, which is used to motivate a wide variety of estimators and tests. Simulation methods, including the bootstrap, are introduced early and used extensively.

The book deals with a large number of modern topics. In addition to bootstrap and Monte Carlo tests, these include sandwich covariance matrix estimators, artificial regressions, estimating functions and the generalized method of moments, indirect inference, and kernel estimation. Every chapter incorporates numerous exercises, some theoretical, some empirical, and many involving simulation.

Econometric Theory and Methods is designed for beginning graduate courses. The book is suitable for both one- and two-term courses at the Masters or Ph.D. level. It can also be used in a final-year undergraduate course for students with sufficient backgrounds in mathematics and statistics.

FEATURES

Unified Approach: New concepts are linked to old ones whenever possible, and the notation is consistent both within and across chapters wherever possible.

Geometry of Ordinary Least Squares: Introduced in Chapter 2, this method provides students with valuable intuition and allows them to avoid a substantial amount of tedious algebra later in the text.

Modern Concepts Introduced Early: These include the bootstrap (Chapter 4), sandwich covariance matrices (Chapter 5), and artificial regressions (Chapter 6).

Inclusive Treatment of Mathematics: Mathematical and statistical concepts are introduced as they are needed, rather than isolated in appendices or introductory chapters not linked to the main body of the text.

Advanced Topics: Among these are models for duration and count data, estimating equations, the method of simulated moments, methods for unbalanced panel data, a variety of unit root and cointegration tests, conditional moment tests, nonnested hypothesis tests, kernel density regression, and kernel regression.

Chapter Exercises: Every chapter offers numerous exercises, all of which have been answered by the authors in the Instructor's Manual. Particularly challenging exercises are starred and their solutions are available at the authors' website, providing a way for instructors and interested students to cover advanced material.

Econometric Theory and Methods provides a unified treatment of modern econometric theory and practical econometric methods. It emphasizes the method of moments and the geometrical approach to least squares. Simulation methods, including the bootstrap, are used extensively, and many of the exercises involve simulation.

Preface

Data, Solutions, and Corrections

1. Regression Models

1.1. Introduction

1.2. Distributions, Densities, and Moments

1.3. The Specification of Regression Models

1.4. Matrix Algebra

1.5. Method-of-Moments Estimation

1.6. Notes on Exercises

1.7. Exercises

2. The Geometry of Linear Regression

2.1. Introduction

2.2. The Geometry of Vector Spaces

2.3. The Geometry of OLS Estimation

2.4. The Frisch-Waugh-Lowell Theorem

2.5. Applications of the FWL Theorem

2.6. Influential Observations and Leverage

2.7. Final Remarks

2.8. Exercises

3. The Statistical Properties of Ordinary Least Squares

3.1. Introduction

3.2. Are OLS Parameter Estimators Unbiased?

3.3. Are OLS Parameter Estimators Consistent?

3.4. The Covariance Matrix of the OLS Parameter Estimates

3.5. Efficiency of the OLS Estimator

3.6. Residuals and Error Terms

3.7. Misspecification of Linear Regression Models

3.8. Measures of Goodness of Fit

3.9. Final Remarks

3.10. Exercises

4. Hypothesis Testing in Linear Regression Models

4.1. Introduction

4.2. Basic Ideas

4.3. Some Common Distractions

4.4. Exact Tests in the Classical Normal Linear Model

4.5. Large-Sample Tests in Linear Regression Models

4.6. Simulation-Based Tests

4.7. The Power of Hypothesis Tests

4.8. Final Remarks

4.9. Exercises

5. Confidence Intervals

5.1. Introduction

5.2. Exact and Asymptotic Confidence Intervals

5.3. Bootstrap Confidence Intervals

5.4. Confidence Regions

5.5. Heteroskedasticity-Consistent Covariance Matrices

5.6. The Delta Method

5.7. Final Remarks

5.8. Exercises

6. Nonlinear Regression

6.1. Introduction

6.2. Method-of-Moments Estimators for Nonlinear Models

6.3. Nonlinear Least Squares

6.4. Computing NLS Estimates

6.5. The Gauss-Newton Regression

6.6. One-Step Estimation

6.7. Hypothesis Testing

6.8. Heteroskedasticity-Robust Tests

6.9. Final Remarks

6.10. Exercises

7. Generalized Least Squares and Related Topics

7.1. Introduction

7.2. The GLS Eliminator

7.3. Computing GLS Estimates

7.4. Feasible Generalized Least Squares

7.5. Heteroskedasticity

7.6. Autoregressive and Moving-Average Processes

7.7. Testing for Serial Correlation

7.8. Estimating Models with Autoregressive Errors

7.9. Specification Testing and Serial Correlation

7.10. Models for Panel Data

7.11. Final Remarks

7.12. Exercises

8. Instrumental Variables Estimation

8.1. Introduction

8.2. Correlation Between Error Terms and Regressors

8.3. Instrumental Variables Estimation

8.4. Finite-Sample Properties of IV Estimators

8.5. Hypothesis Testing

8.6. Testing Overidentifying Restrictions

8.7. Durbin-Wu-Hausman Tests

8.8. Bootstrap Tests

8.9. IV Estimation of Nonlinear Models

8.10. Final Remarks

8.11. Exercises

9. The Generalized Methods of Moments

9.1. Introduction

9.2. GMM Estimators for Linear Regression Models

9.3. HAC Covariance Matrix Estimation

9.4. Tests Based on the GMM Criterion Function

9.5. GMM Estimators for Nonlinear Models

9.6. The Method of Simulated Moments

9.7. Final Remarks

9.8. Exercises

10. The Method of Maximum Likelihood

10.1. Introduction

10.2. Basic Concepts of Maximum Likelihood Estimation

10.3. Asymptotic Propertied of ML Estimators

10.4. The Covariance Matrix of the ML Estimator

10.5. Hypothesis Testing

10.6. The Asymptotic Theory of the Three Classical Tests

10.7. ML Estimation of Models with Autoregressive Errors

10.8. Transformations of the Dependent Variable

10.9. Final Remarks

10.10. Exercises

11. Discrete and Limited Dependent Variables

11.1. Introduction

11.2. Binary Response Models: Estimation

11.3. Binary Response Models: Inference

11.4. Models for More than Two Discrete Responses

11.5. Models for Count Data

11.6. Models for Censored and Truncated Data

11.7. Sample Selectivity

11.8. Duration Models

11.9. Final Remarks

11.10. Exercises

12. Multivariate Models

12.1. Introduction

12.2. Seemingly Unrelated Linear Regressions

12.3. Systems of Nonlinear Regressions

12.4. Linear Simultaneous Equations Models

12.5. Maximum Likelihood Estimation

12.6. Nonlinear Simultaneous Equations Models

12.7. Final Remarks

12.8. Appendix: Detailed Results on FIML and LIML

12.9. Exercises

13. Methods for Stationary Time-Series Data

13.1. Introduction

13.2. Autoregressive and Moving-Average Processes

13.3. Estimating AR, MA, and ARMA Models

13.4. Single-Equation Dynamic Models

13.5. Seasonality

13.6. Autoregressive Conditional Heteroskedasticity

13.7. Vector Autoregression

13.8. Final Remarks

13.9. Exercises

14. Unit Roots and Cointegration

14.1. Exercises

14.2. Random Walks and Unit Roots

14.3. Unit Root Tests

14.4. Serial Correlation and Unit Root Tests

14.5. Cointegration

14.6. Testing for Cointegration

14.7. Final Remarks

14.8. Exercises

15. Testing the Specification of Econometric Methods

15.1. Introduction

15.2. Specification Tests Based on Artificial Regressions

15.3. Nonnested Hypothesis Tests

15.4. Model Selection Based on Information Criteria

15.5. Nonparametric Estimation

15.6. Final Remarks

15.7. Appendix: Test Regressors in Artificial Regressions

15.8. Exercises

References

Author Index

Subject Index