on 11 June 2012
This book was pretty much the only reference book I used for Econometrics in 2nd and 3rd year economics, and a useful guide for some of the applied courses for Masters courses.
A couple of reasons why it is excellent:
1) The book doesn't really presume any knowledge about any of the topics it covers. Therefore, don't be afraid that you'll need additional primers to make sense of it, beyond a very basic understanding of hypothesis testing and statistics
2) There is excellent description of the techniques and maths behind the processes, as well as quite well thought through examples that illuminate the difficulties of using the technique, without alienating anyone who isn't as strong in maths.
3) It covers a lot of material, both in cross sectional and time-series econometrics (although time series is a bit weaker)
4) Its really aimed at 'practical' uses of econometrics, the kinds of things a typical economist will use to test hypotheses in dissertations etc. The focus is on helping you start up a program, run a regression and interpret the results correctly - not explaining what the program will be doing and how the regression is determined.
What its less good at:
1) The hardcore mathematics underpinning the processes - The book does give it a try for some of the simpler stuff, but don't expect a full proof of what OLS is doing
2) Being succinct - The book does sometimes hide key facts behind a lot of words. Example: Classical errors in variables testing - although subheaded, the real key points are sometimes lost in a lot of text, which can mean its difficult to revise from quickly
In summary - this book is excellent for anyone starting out with Econometrics. My copy has been in constant use ever since I got it!
on 17 October 2012
I wanted this book for a long time, I got it because my husband started a second degree and he has to do econometrics now. I have the "Econometric Analysis of Cross Section and Panel Data" and "Introductory Econometrics" is a little sister to it. I am not exaggerating, but I have not found other econometrics book, which combines the required "econometric" jargon with intuition. Through the whole book no concept is left without an example or an application, which is really good to help you understand what is going on. Worth every penny.
On the lookout for a suitable text with which to brush up on my very limited and very brief study of econometrics, this was one of the ones recommended to me by a relative, an economics professor at Wilfrid Laurier University in Ontario.
In its prime purpose, the aforementioned revision, Wooldridge's book more than succeeds, in not only explaining how to perform various econometric estimations but also how they work, the underlying mathematics, some of the associated problems, the best applications for them, and how to interpret results. As such I'd have no hesitation in recommending the book myself to anyone wishing to start or refresh their knowledge of the subject. But I'd also recommend for some reading the Appendices first, as these provide the mathematical grounding for much of what is in the main text.
One of the chief selling points for me was the availability of lots of datasets with which to work. Unfortunately one of the assumptions of the author or publishers is that readers are full-time students at institutions whose teachers have answer books. As this was not the case for me I contented myself with at least being able to match the computed answers in examples in the text.
When I initially downloaded the datasets I was puzzled by the absence of headings in the Excel spreadsheets. These, I was told, were contained separately in associated text files (why?). Fortunately I was spared the manual grind of pasting the headers when I downloaded the software I was going to use, open source and free gretl, where I found there were already gretl-friendly files of Wooldridge's datasets on the home site. At that point I felt glad I was unable to afford a more expensive package.
Wooldridge's explanations are careful and meticulous, and he goes out of his way to avoid ambiguity, which sometimes makes reading a little like going through a legal document, but is necessary given the sometimes complex nature of his subject. (There is at least one occasion, though, where the complexity of one sentence gets the better of him and he loses it himself.) His style could, though, be a little more concise if he did not insist on personalising everything ("we" find this, "we" need to do that) or offering little false comforters on occasion (resolving "this" is quite straightforward, or easy, or no problem).
Lacking easy access to tutors or anyone with an econometrics background (I live five timezones away from Ontario), sometimes I encountered problems, some of which I was unable to resolve on my own. With Example 8.1 I tried several times to match the author's result with no success. In Example 9.1 I managed to work out that a couple of numbers had been transposed in the text, hence why my results did not match. The data in Example 9.4, amongst others, needs so much background manipulation I didn't bother. In example 10.6, again run several times, my results came close, but were still not the same, and again, in Example 14.4 my results for Pooled OLS and Random Effects were close, but for Fixed Effects several worlds away. And finally, in the more advanced sections, I hit problems such as those in Example 17.5 where gretl asked for Selection Variable, Regressors and Selection Regressors, but there was no mention of these beasties in the text.
However, whilst buyers of the book may like to know it isn't perfect, a handful of glitches in a book of over 800 pages isn't bad, and certainly don't detract from its overall value.
on 31 December 2012
When I first started studying econometrics a few years ago I thought very little about this book, it was just a book which the courses tended to be structured around. However a few years on, I now understand that the reason why my introductory course (and virtually every introductory course in every decent university in the world) was structured around chapters 1-7 of this book is the fact that these chapters have become the basis of modern econometrics. Before Wooldridge were ideas and concepts, many (Gujarati etc) tried to put them all in one place, but the terminology and structure of Wooldridge has become the single text that the subject is taught based on, with its unique terminology and presentation of formulas becoming the 'norm'. This book is essential for any economics / econometrics student: you simply can't do your degree without delving into Wooldridge.
I've given it four stars, as I personally found that whilst it is great as a referencing text it does not introduce the subject sufficiently such that it could be used as more than a supplement to a closely taught course. Rather, I think the title is a might misleading as works such as Understanding Econometrics (Stewart, 1984) do a far better job of introducing the subject. Also for an allegedly introductory text, much of the notation will put off those from non-mathematical backgrounds, perhaps unnecessarily so, as if explained more with prose students would see some notation for the simple concepts they are. One other criticism is the way the F-Test formula is constructed can cause some confusion later on into studying when applying it to cross sectional data: there are more generic ways of presenting it that would ease learning for students.