TITLE : BASIC ECONOMETRICS. 3RD ED.

In this edition I have attempted to incorporate some of the developments in the theory and practice of econometrics that have taken place since the publication of the second edition in 1988. In addition, this revision has given me the opportunity to simplify the discussion of some topics included in the previous editions and to add some new material on these topics. The major changes in this edition are as follows:

1. In Chapter 1, I have expanded the discussion of the nature and sources of data available for econometric analysis. In view of the increasing use of time series data in economic analysis, I have introduced very early the concept of a stationary TIME series, a concept that is crucial for analyzing data involving economic time series.

2. In Chapter 3, I present a more extended discussion of the assumptions of the classical linear regression model (CLRM). The CLRM is the foundation of econometrics. In this chapter I also discuss the Monte Carlo simulation experiments.

3. In Chapter 5, on hypothesis testing, I have introduced the concept of the p value, or the exact level of significance, of a test statistic. In this chapter, I also discuss the Jarque-Bera test of normality.

4. In Chapter 8, on hypothesis testing in the context of multiple regression models, I have streamlined the discussion. This chapter also includes a discussion of the choice between linear and log-linear regression models. In the appendix to this chapter, I discuss, at an elementary level, the likelihood ratio (LR) test of hypothesis.

5. In Chapter 10, on multicollinearity, I now give equal billing to niicronumerosity (smallness of sample size), a concept due to Arthur Gold berger. I also introduce tolerance and inflation-variance tools for detecting multicollinearity.

6. In Chapter 11, on heteroscedasticity, I have now included the BreuschPagan-Godfrey test and White's test of heteroscedasticity. I also discuss White's heteroscedasticity-consistent variances and standard errors of OLS estimators.

7. In Chapter 12, on autocorrelation, I have included these tests: asymptotic test of autocorrelation, Breusch-Godfrey test of higher-order autocorrelation, and Berenblut-Webb test. Included in this chapter is also the ARCH model, which has been increasingly'used in financial economics.

8. Chapter 13, on model building, discusses nominal versus true level of significance in the presence of data mining, and the Lagrange multiplier (LM) test for adding variables to a regression model.

9. In Chapter 14, which is new, I discuss alternatives to the traditional econometric methodology. In particular, I discuss Leamer's and Hendry's approaches to econometrics. Also included in this chapter are tests of non-nested hypotheses, in particular the Davidson-MacKinnon J test.

10. Chapter 15, on dummy variables, now includes a discussion of dummies in combining time series and cross-sectional data. I also show how the dummies can be used in the presence of autocorrelation and heteroscedasticity. An exercise in this chapter discusses Zellner's seemingly unrelated regression (SURE) technique.

11. Chapter 16, on dummy dependent variable regression models, now includes a discussion of the Tobit model.

12. Chapter 17, on dynamic regression models, now includes a discussion of both the Granger test and Sims's test of causality.

13. Chapters 18, 19, and 20, on simultaneous-equation models, now contain tests of simultaneity and exogeneity. These chapters also discuss the relationship between causality and exogeneity.

14. In recognition of the growing importance of time series data in economic analysis, I have included two new chapters on time series econometrics. In Chapter 21, I introduce the key concepts of time series analysis, such as stationarity, random walk, unit root, Dickey-Fuller and augmented Dickey-Fuller tests of stationarity, deterministic and stochastic trends, trend-stationary and difference-stationary stochastic processes, cointegration, Engle-Granger tests of cointegration, error correction mechanism, and spurious regression. In Chapter 22 I discuss the Box-Jenkins, or ARIMA, and vector autoregression (VAR) approaches to economic forecasting. These are alternatives to the traditional single- and simultaneous-equation approaches to forecasting.

I have added several new exercises. The exercises given at the end of chapters are now divided into two groups: questions and problems. The latter are data-based exercises (I am a firm believer in learning by doing).

All these changes have considerably expanded the scope of this book. I hope this gives the instructor substantial flexibility in choosing topics that are appropriate to the intended audience. Here are some suggestions about how this book may be used: One-semester course for the nonspecialist: Appendix A; Chaps. 1 through 8; an overview of Chaps. 10, 11, and 12 (omitting all the proofs); and Chap. 15. The theoretical exercises can be omitted. One-semester course for economics majors: Appendix A; Chaps. 1 through 8; and Chaps. 10 through 15. If matrix algebra is used, include Appendix B and Chap. 9. Some of the theoretical exercises can be omitted. Two-semester course for economics majors: Appendices A and B, and Chaps. 1 through 22. Mathematical proofs given in the various appendices can be covered on a selective basis. Additionally, the instructor may want to cover the topic of nonlinear (in parameters) regression models.

This revision would not have been possible without the constructive comments, suggestions, and encouragement that I have received from several people who have read the various drafts. In particular, I would like to acknowledge my debt to the following professors, without, of course, holding them responsible for any deficiencies that remain in the book: Ted Amato, University of North Carolina; Dale Belman, University of Wisconsin-Milwaukee; Tom Daula, U.S. Military Academy; Mary Deily, Lehigh University; Frank Diebold, University of Pennsylvania; David Garman, Tufts University; Sushila GidwaniBushchi, Manhattan College; William Greene, New York University; Dennis Jansen, Texas A&M University; Jane Lillydahl, University of Colorado; Dagmar Rajagopal, Ryerson Polytechnic University; Bo Ruck, U.S. Military Academy; John Spitzer, State University of New York, Brockport; and H. D. Vinod, Fordham University.

I owe a great debt of gratitude to Professor Kenneth J. White and Steven A. Theobald for checking numerical calculations and for preparing the handbook: Basic Econometrics: A Computer Handbook Using SHAZAM, 3d ed., McGrawHill, New York, 1995. My wife Pushpa and my daughters Joan and Diane have been a constant source of inspiration and encouragement. Beyond a simple thank-you, I want them to know that I love them all very dearly.

On a personal note, after some 28 years of teaching at the City University of New York (CUNY), I have now joined the Department of Social Sciences at the U.S. Military Academy at West Point, New York. I am grateful to CUNY for providing me with my first job and to the Academy for offering me new challenges and opportunities.

Damodar N. Guiarati