Econometrics

Econometrics has many useful features and covers all the important topics in econometrics in a succinct manner. All the estimation techniques that could possibly be taught in a first-year graduate course, except maximum likelihood, are treated as special cases of GMM (generalized methods of moments). Maximum likelihood estimators for a variety of models (such as probit and tobit) are collected in a separate chapter. This arrangement enables students to learn various estimation techniques in an efficient manner. Eight of the ten chapters include a serious empirical application drawn from labor economics, industrial organization, domestic and international finance, and macroeconomics. These empirical exercises at the end of each chapter provide students a hands-on experience applying the techniques covered in the chapter. The exposition is rigorous yet accessible to students who have a working knowledge of very basic linear algebra and probability theory. All the results are stated as propositions, so that students can see the points of the discussion and also the conditions under which those results hold. Most propositions are proved in the text.

For those who intend to write a thesis on applied topics, the empirical applications of the book are a good way to learn how to conduct empirical research. For the theoretically inclined, the no-compromise treatment of the basic techniques is a good preparation for more advanced theory courses.

Unlike the vast majority of econometrics texts, this book combines solid economic theory with a thorough grounding in basic mathematical statistics. The worked practical examples introduce some classic empirical papers and provide an excellent motivation to study the theory. The little bit of effort required to put on the GMM-tinged glasses makes everything look so much clearer.

The incidental treatment of linear algebra is better than that of any 'Math. for Economists' book I have ever come across.

Anyone at the graduate level of study should own this book. It would also be of value to advanced undergraduates, and out-of-touch academics.

It should be noted that an important aspect of statistical analysis is the treatment of 'outliers', such as the 'review' presented by Mr. Brian J. Phillips.


Solid basis for econometric analysis   April 2, 2006
Phome (NY, USA)
25 out of 25 found this review helpful

I have a love/hate relationship with this book. Perhaps I should state as a precursor that I was never formally trained in economics before learning econometrics. And, that the last time I'd done matrix algebra or calculus was some 15 years prior.

We used this book as part of a taught graduate course. It took half a semester to go through the first two chapters - an investment of time that proved well worth it for the rest of the topics which were covered in the remainder of the semester.

Basically, if you can understand the first two chapters on ordinary least square regression for finite and large samples, the required assumptions and properties, then the rest of the chapters are a piece of cake:
- generalized method of moments for single and multiple equations
- panel data
- time series analysis (including unit root analysis)
- extremum estimators
- maximum likelihood
- cointegration.

In short, the book covers all major econometrics topics and does so in a succinct, clear manner. The way in which Hayashi builds on each topic, showing that all models are basically different versions of the same method, with slightly different assumptions is just brilliant. It put statistics in a different light for me, and gave me a much deeper, intuitive understanding of it than any other book or class had done before.

There is a caveat however. This book assumes that you have substantial mathematical grounding. In particular, I found the succinct use of notation, without any verbal explanation, irritating at first. I invested quite some time in a mathematical economics book reminding myself what sets were, rules of matrices, calculus functions, expectations and probability.

Without the support and input of our brilliant teacher who (very patiently) took us through the end of chapter exercises step-by-step, I would never have managed to successfully read this book on my own! While those exercises honed my skills and deepened my understanding, I relied heavily on Hayashi's home page notes and hints to complete them.

For those of you that have strong mathematical skills and an economic background, this book is probably one of the best introductions to econometrics. For those of you who do not, it will prove to be a difficult read at best.

What's certain is that after succesfully completing it, your econometrics and statistical skills will provide a solid enough basis for any graduate program.






The Most Readable Econometrics Text There Is. Period   October 18, 2004
jose
22 out of 23 found this review helpful

I think Hayashi is the best econometrics textbook to come along in a long time. The treatment has that rare quality of being simultaneously sophisticated yet very easy to follow. In that sense, this book is much different than Greene - whereas Greene is (I think) much more of a reference, you can actually sit down and learn a lot of econometrics with this book. Hayashi not only takes the time to explain key concepts in good prose, but in some cases even writes down step-by-step instructions. All this while not compromising the material.

The treatment is also slightly different in that GMM is a central theme instead of something off to the side, which is very nice. There are plenty of empirical examples - these are somewhat helpful, and the exercises are fairly easy but still illustrative.

Two downsides - it would have been nice to see some treatment of Bayesian econometrics, since this appears to be used much more widely (Lancaster is a good supplement). Second, either I got a faulty book, or there are no tables of critical values. This is ultimately a minor gripe since just about every other book has tables (and you really don't even need them these days with packages and such), but it can be annoying.

Ultimately, the combination of sophistication and readability of this book is what sets it apart from all others. If you're looking to learn econometrics, buy this book.


Excellent   March 20, 2003
Ruben Osuna Guerrero (Madrid Espana)
14 out of 16 found this review helpful

This book has excellent qualities:
1) Its clarity and concision, in exposition and proofs.
2) The modern approach and well structured and complete contents
3) The empirical exercises

But, for pure theory I prefer Davidson & McKinnon's "Estimation and inference", and for an empirical approach Berndt's "Practice of Econometrics".

Hayashi's is a good intermediate alternative, with great theory exposition and good empirical exercises.

One of the best options for a graduate student.


A modern and unusal approach   January 27, 2005
Sabad One
12 out of 14 found this review helpful

This is a fine book, but probably not the one you want to buy if you are looking for ONE all-encompassing reference. The approach is interesting, but unusual, with all the pros and cons that come with originality. It is for graduate students, or very advanced undergraduates, as it requires quite a lot of previous knowledge of linear algebra and statistics.
What is unusual about this book is that it covers most topics within a unifying Generalized Method of Moments (GMM) framework. Many many estimators are treated as special cases of GMM. The book is clear, and the notation is mostly OK, even if the chapters on panel data and systems of simultaneous equations are a notational nightmare, partly because of the choice of treating everything in a GMM framework. Another unusual aspect of this book is the emphasis on certain regularity conditions (such as ergodicity) that are usually used in a time-series framework, but are not commonly seen in cross-section analysis. I studied (also) on this book as a graduate student, and overall I liked it. The only real minus are the exercises, which contain so many hints that they become trivial (really, I am not a genius...). Worse, they only require mindless application of linear algebra.

One UNimporant cons of this book is the fact that (cover aside) it is... ugly! How could the publisher choose the boring "Times New Roman" font for this book!? But this, of course, does not really matter....

Overall, a useful and good book, but if you are looking for ONE textbook in cross-section econometrics Wooldridge is probably a better choice, and if what you are looking for is ONE book in time series, Hamilton is likely to be what you want on your shelf.