Designed for a first course in introductory econometrics, Introduction to Econometrics, reflects modern theory and practice, with interesting applications that motivate and match up with the theory to ensure students grasp the relevance of econometrics. Authors James H. Stock and Mark W. Watson integrate real-world questions and data into the development of the theory, with serious treatment of the substantive findings of the resulting empirical analysis.
It's not good. If you already understand what's going on, it seems like a decent textbook, but Stock & Watson often lack clarity in their explanations. This book is 1) nearly impossible to skim because of the writing style 2) very bad in terms of making connections to other theories and applications and ideas 3) devoid of metaphor and 4) bad at concisely explaining equations. A good mathematical textbook gives you an important equation and then explains it for the next paragraph, telling you what all the variables mean, what happens if some of them change, what this means, etc. S&W often don't even tell you what variables mean, and you might have to go three pages back in the book to figure out that the variable p is supposed to stand for "the number of lags." And the amount of times they say something like "and we can prove this because it's just like equation (8.7)," and then don't even mention what equation 8.7 is infuriating. They expect you to go back 200 pages to equation 8.7 (which, by the way, has no concise explanation, so you're going to need to read the five pages around equation 8.7 to figure out what it means). It's like reading a research paper that leaves page numbers without quotations--it treats me like I'm too stupid to do the thinking myself.
Although I've nothing to compare it against, the book lacks usability. Too much of the rendering of equations is written in block text rather than using any diagrams. There are other points where equations are re-render/substituted/expanded/simplified without explanation. One is left to figure out these leaps on his own. That's not instruction or elucidation. Typical academic fair whereby the author is too far removed from his audience (an introductory one) so that much is lost.
This is a decent undergraduate econometrics textbook, but I think that "Introductory Econometrics: A Modern Approach" by Jeffrey Wooldridge is a much better choice for learning econometrics. For this book, the Introduction to Econometrics website provides some useful links and resources.
This book is appropriate as a gentle introduction to econometrics (the application of statistical methods to economic theory). The book is well-written and the authors focus on insuring that the book does not fly over the heads of beginners.
Stock's Introduction to Econometrics is probably the best undergrad text I've found on regression, explaining even some fairly advanced topics in an accessible way. I've used it extensively to try to understand some of the hairier concepts I've been dealing with in my first-year grad sequence.
People who have a knack for statistics tend to not have one for explaining statistics, but this book is one of the most accessible textbooks out there. Thorough but readable.