Customer Reviews

Mostly Harmless Econometrics: An Empiricist's Companion

5 star:		(11)
4 star:		(4)
3 star:		(2)
2 star:		(3)
1 star:		(0)

 

 

This is a wonderful book. Despite having taken many courses and read many statistics and econometrics books, I'm sometimes stuck in my own applied problems. Of course, these courses and books taught me how to derive asymptotic theorems or be careful when maximum likelihood fails, but this is not what I really need to know to solve my problems. Now with Angrist and Pischke, I have a book that is truly applied in focus -- one that explains why and how certain empirical strategies are convincing and one that is up-to-date with the latest examples of recent research employing these strategies. Real econometrics applied to real problems. This book should be on every applied economists bookshelf.

In their introduction Angrist and Pischke say "empirical research is most valuable when it uses data to answer specific causal questions, as if in a randomized clinical trial". The book aims to teach this approach to students of applied econometrics.

The emphasis on natural experiments and quasi-experiments which the authors espouse has become influential in some sub-areas of econometrics and the authors, particularly Angrist, have played a leading role in this development. However, this approach is not uncontroversial. The Journal of Economic Perspectives has an entire issue (Spring 2010, Vol. 24, No. 2, full text online for public) discussing the pros and cons and you may want to glance through it before buying this book.

Taken on its own terms the book suffers from indiscriminately mixing four levels of discussion: the philosophical and methodological issues around causality, tips and tricks on how the apply the workhorse models of (micro)econometrics, case studies, and the mathematical properties of models and estimators. The authors keep switching constantly between these four levels rather than presenting them sequentially, making the book very hard to read and follow. The problem is compounded by the sketchiness of the mathematical derivations and the lack of a precise definition of the key idea of 'causality'.

If you are starting out in econometrics you may be better served by traditional textbooks with more detailed and careful presentations such as Wooldridge's Introductory Econometrics and Econometric Analysis of Cross Section and Panel Data.

If your interest is in causal inference, deeper treatments can be found in Judea Pearl's Causality: Models, Reasoning and Inference or Morgan and Winship's Counterfactuals and Causal Inference.

This is a great book that I really wish had been written when I was first learning econometrics. It is probably best appreciated by someone who has already taken one or two econometrics classes and has had some exposure to applied econometrics research. If you haven't taken econometrics yet you will not really appreciate the 'paradigm shift' that they are trying to explain nor appreciate some of the humor and storytelling. But if you are like myself someone who first learned econometrics more than a decade ago you'll find yourself reading this book page by page with appreciation. I feel that it is helping me to develop a much more mature understanding of the field of applied micro econometrics (helping me to confirm hunches and general notions I already and also new gems of insight). There are many cookbooks that might explain the methods in more depth (e.g. Cameron and Trivedi Microeconometrics) but that's not the purpose of this book. It's purpose is to make you think more about 'research design' which is to say about the questions that you pose, and how you pose them, rather than the methods that you use to try to arrive at answers. For too long econometricians got lost in the details of methods without taking a step back to think about some really fundamental questions first. The book is filled with examples of both good and bad research and you'll be surprised at how bad some past very influential research looks in light of modern day paradigms. It's not that these earlier researchers didn't know enough math, it's that they used the math without clear enough purpose.

This book will make you a better economist and beyond that make you see the world around you slightly differently. You'll end up with a keener eye for all those natural experiments happening all around you. If you are an advanced undergraduate, MA or starting PhD student with any pretensions of becoming engaged in original applied economics research this book should be a "must have" on your bookshelf.

The first thing I want to say is this: If you plan on doing regression analysis in your research, stop what you are doing, and read this book first. I think this book represents THE current statement on how we should use regression. For Angrist and Pischke, regression is a technology for summarizing data. If regression is to be used for causal inference, then there is nothing in the specification of the model or the choice of estimator that can ultimately make the causal story persuasive. That is, you don't identify causal effects simply by including "control" variables in your regression. The identification comes from elsewhere---either a real or "quasi" experiment---and the regression is what you use to clean up the imperfections of the experiment and measure effects. Angrist and Pischke have done an enormous service to social science by writing a regression textbook that nonetheless emphasizes the primacy of design. This is a terrific corrective for the "101 flavors of regression" approach of textbooks to date.

Even with this emphasis on design, Angrist and Pischke show us that are a lot of nuances to the way that regressions measure such effects---e.g., in the presence of effect heterogeneity---and that's what this book explores in exquisite detail. It's a hugely important book and a very serious and rigorous treatment, despite it's apparently causal style. They make some claims that may strike some as outrageous---e.g., always using OLS, even for limited dependent variables---but the rigor of their presentation means that the onus is on those who disagree to think harder about why, exactly, they would prefer, say, a more parametric approach.

Nonetheless, it isn't a "5 star" book. It often feels a bit rough-draft-like. The presentation of technical material skips important steps rather haphazardly. I wonder if this was due to bad editing? Hopefully there will be a second edition that cleans up these rough edges, in which case it would be the ideal textbook on regression analysis.

Amazing book and very helpful, but a terrible idea to buy this on Kindle. The graphs and variables are all formatted strangely, making it hard to read.

After drowning in Heckman, cursing Rubin and struggling with Manski, this is a welcome and needed relief. I think back to what "Thinking Strategically" was to Game Theory and must conclude that having fun must be regular and necessary phase in the gestation of an idea.
The authors take delight in the subject and dole out some of the most interesting applications from a field that's a testament to soporifics.
Trust me...if it weren't for Monster Energy drinks, none of us would understand this stuff.
Their mathematics is approachable by anyone who knows what an expectation is. They take on such opaque subjects as instrumental variables and differences-of-differences with some examples that I promise will stay with you even after you've put the book down. The subjects connect with a real ease and you're never left referring back to earlier chapters or other texts to "remember" something they've assumed you know.
On whole, they approach the subject with a joy I haven't seen since I saw a couple of otters on a water slide on the side of an island in Alaska.

The book is very good. However, I have strong complaints about the Kindle edition: fonts of math characters are very bad. Math characters are not uniform, many of them are much bigger than the text characters but others have the same size as text characters which means that the problem is with a sloppy edition and not a consequence of bad software. Additionally, zooming only works for some of the math characters. I have many other books in pdf format in my Kindle and they work much much better than those I have in the Kindle format.

This is one of the best books on applied econometrics. Unfortunately, the Kindle edition is absolutely slopy, the math characters are weird and blurred. I wish amazon would have been more careful with this Kindle edition. Taking into account that I need this book for my daily job, it seems that I will have to pay for the printed edition since the math characters in the Kindle are barely readable.

I'm not sure what book the other people were reading, but this book was hardly clear and easy to understand. I've taken a graduate Econometrics class and several others regarding statistics, and this book is not for the lay reader to understand econometrics. It is written in a very confusing manner and is not an entry level book. Maybe if you were a PHD candidate or have a MS in mathmatics you would understand it better. I would buy a used Econometrics textbook before buying this one.

I try not to buy more textbooks, but I highly recommend this book. It has helped me to have a systematic approach to research design, which makes the rest of the research process much more efficient. I'll keep this one.