Chapter 1 now includes a discussion of different types of data: experimental, observational, and sample survey. Because most of econometrics deals with modeling one or more variables for given values of other variables that influence the former, the theory part of Chapters 2 and 3 have undergone the most changes. In Chapter 2, the section on joint probabilities has an extensive discussion of conditional probability, conditional expectation, and conditional variance. Properties involving multivariate distributions have been moved from the appendix to the chapter, but these sections are marked with an asterisk so that they can be skipped, if desired, without loss of continuity. In the theoretical part of Chapter 3, all the assumptions on the error terms have been stated as conditional on given X. Proofs of unbiasedness and the Gauss-Markov Theorem are initially done with X given and then extended to the case where it is a random variable. Chapter 4 has a new section on generating the forecast and confidence interval for the dependent variable in a multiple regression model. Also added here is a section on interpreting regression coefficients, especially those for variables measured as percentages or proportions. This emphasis on the proper interpretation of the results has been re-inforced and extended in Chapter 6 with a discussion and a table summarizing the interpretation of regression coefficients in models involving logarithms of variables. Chapter 7 presents modified interpretation of dummy coefficients in the case of log-linear models. Chapter 10 has also undergone substantial changes with new sections on Vector Autoregressive (VAR) models and the treatment of panel data. Chapter 11 extends the forecast combination section to include serial correlation, ARCH effects, and time-varying weights. In Chapter 14 on the steps for carrying out empirical projects, there is now an extensive discussion of how index numbers are computed and the importance of dealing appropriately with changes in the base periods.
Several new data sets, end of chapter exercises, and references have been added. All the data are now available in a variety of formats; ASCII (that is, text format), B34S, EVIEWS, EXCEL, GRETL, PcGive, and SHAZAM. Appendix D has a list of the data with the names of the variables, units of measurement, and the data sources. The instructor's manual has also been revised to include answers for the new exercises.
People familiar with the fourth edition of the book are aware of the econometrics programs ESL/ESLWIN included with the book. Professor Allin Cottrell has spent numerous months developing GRETL (Gnu Regression, Econometric, and Time-series Library), the successor to his ESLWIN, which is now a completely Windows-based and more user-friendly program with numerous additional econometric methods, and has graciously consented to include it with this book. There are not enough thanks to express my deepest appreciation of his effort and generosity in giving it free to the world. All the Computer Practice Sessions set up for ESL/ESLWIN have been redone to be accessible by GRETL. The program, its manual, the book's data sets in different formats, ready-made commands for practice sessions, and GRETL databases, are all included in the CD accompanying this book. A brief introduction to the program is provided in the link to Practice Computer Sessions.
Many professors expressed annoyance at the interruption of the flow of the text material by computer outputs with commands tailored to the ESL program provided with the previous edition of the book. To accommodate this complaint, I have now deleted all references to commands and kept only essential outputs, that are basically independent of the software that readers use, so that the comments in these outputs will be beneficial to all. This, combined with the new feature of having data sets readily readable by a variety of econometric packages, should make the task of empirical analysis considerably easier. The text references to Practice Computer Session files such as PS2-1.INP can be ignored by nonGRETL users or, better yet, the commands may be used as a guide to reproducing the examples using their own regression package.
I would like to acknowledge with gratitude the kindness of Professor Ken White and his colleagues for accessibility to SHAZAM and for preparing a supplementary disk with the program's commands for reproducing all the book's examples, Professors David Hendry and Jurge Doornik for setting up the book's data files in a form accessible by the PcGive program, and Professor Houston Stokes for the same by the B34S program.
I am deeply indebted to all the instructors who have used the previous editions of the book and to the numerous students who compiled many of the data sets include here (their contribution is acknowledged in Appendix D). I am also greatly thankful to a number of professors who made extensive comments on the manuscript. They are, David Brasington (Tulane University), Anthony M. Carilli (Hampden-Sydney College), Allin Cottrell (Wake Forest University), Graham Elliot (University of California, San Diego), Lily Huang (University of California, San Diego), Oscar Jorda (University of California, Davis), Jan Kiviet (University of Amsterdam), Judy Mann (University of California, San Diego), Brad Tuck (University of Alaska), H. van Ophem (University of Amsterdam), and Phanindra V. Wunnava (Middlebury College). Most of their page by page comments have been incorporated, but some of the suggestions for drastic changes could not be accommodated because of a pressure on time. I hope to address them through supplementary pages in the Website set up for this book. Many thanks are also due to Gail Gavin, the Project Editor, Angela Urquhart, the Project Manager, and to Janet Hennies, the Marketing Strategist, for making the production and marketing processes smooth. Also deserving thanks are Lynn Reichel, Randy Campbell and Ayse Pinar Tutus who painstakingly proofread all aspects of the book. Finally, I would like to acknowledge the help from Amy Porubsky of Harcourt College Publishers for the years of dedication towards helping me in a variety of ways, particularly with reviews of the fourth edition, requests from professors for the instructor's manual, and so on.
I welcome comments from readers on any aspect of the book (email address: firstname.lastname@example.org). I would be particularly grateful to users who point out typographic errors that invariably crop up even after scrupulous proof-reading.