以下是引用shikelang在2008-10-6 22:20:00的发言:
Statistical Methods in Econometrics
By Ramu Ramanathan
Publisher: Academic Press Number Of Pages: 405 Publication Date: 1993-01-07 ISBN-10 / ASIN: 0125768303 ISBN-13 / EAN: 9780125768306 Binding: Hardcover
Product Description:
Statistical Methods in Econometrics is appropriate for beginning graduate courses in mathematical statistics and econometrics in which the foundations of probability and statistical theory are developed for application to econometric methodology. Because econometrics generally requires the study of several unknown parameters, emphasis is placed on estimation and hypothesis testing involving several parameters. Accordingly, special attention is paid to the multivariate normal and the distribution of quadratic forms. Lagrange multiplier tests are discussed in considerable detail, along with the traditional likelihood ration and Wald tests. Characteristic functions and their properties are fully exploited. Also asymptotic distribution theory, usually given only cursory treatment, is discussed in detail.
The book assumes a working knowledge of advanced calculus (including integral calculus) basic probability and statistics, and linear algebra. Important properties from matrix algebra are summarized in the appendix. Numerous examples, exercises, and practice problems are included.
Key Features
* Covers both multivariate analysis and matrix algebra
* Focuses on newer tests of hypotheses such as the Lagrange multiplier test
* Discusses characteristic functions in depth
* Material has evolved during 15 years of classroom instruction
Summary: Outstanding Reference!
Rating: 5
Simply put, this book has to be the best general reference for econometrics that I have ever come across. I bought this book after reading about it and because of a good experience with another of the author's texts. I only wish I had gotten it sooner!
This book is well organized and covers a tremendous amount of econometric theory and results. It never failed me that when I was unclear on a subject or needed clarification, this book provided me with clear insight. It is an outstanding companion to many of the "headier" econometrics texts out there, particularly those used in graduate classes (Davidson/McKinnon, Greene, etc).
Please, do yourself a favor and splurge on this book. You will not regret it and you'll probably help your grades at the same time.
Summary: Very, very useful book
Rating: 5
When I first started grad school, I was pretty lost when it came to econometrics. I wondered why things were so difficult. Then I found out that there was a large gap between the statistics I knew, and the statistics I needed to know before starting econometrics. I read this book on my own, and it quickly taught me the stats I needed to know. It's very good for self study, since there are many worked out examples. The book is also relatively short, and is geared to get people up and running quickly. It seems designed for people in my situation. It's a prep course for graduate level econometrics. If you have done only basic statistics and econometrics in undergrad, definitely work through this stuff before you start grad school and apply the probability and statistics in this book to econometric models, especially if your grad dept, like mine, wrongly assumes that everybody knows this already and delves right into econometrics.
Summary: Good for quick review
Rating: 5
If you want to have a book for quick reference and review on basic statistics, this is the one.
It is very well written, but it does not give extensive explanations on the topics it covers, because it assumes the reader has already studied the subject.
I think this is the perfect book for a Summer course, usually given to entrants PhD students in Economics.
All the important results that a PhD student in Economics is going to use are there, either derived in the body of the text, or as an exercise. By the way, the exercises are very well posed, they present very interesting results, and the difficulty level is pretty appropriate.
The high points are Part I - Probability theory, and Part II - Statistical Inference. In the former, generating functions and all relevant distributions in Statistics are discussed, as well as the mathematical relationship between the distributions. In the latter, the instructor should complete the asymptotic distribution theory with another reference. Part III - Econometrics - should be covered in a real Econometrics course, although it talks about heteroscedaticity, surprisingly including GARCH models.