Book review: Mastering 'Metrics

Last year, I reviewed Mostly Harmless Econometrics, by Joshua Angrist and Jorn-Steffen Pischke, where I noted that:

...this is a book that is not for the fainthearted undergraduate economics student...

Fortunately, Angrist and Pischke have a more accessible book that I just finished reading, Mastering 'Metrics. This book is everything for undergraduates that Mostly Harmless Econometrics is for graduate students, and more. I found it incredibly accessible and readable, which sets it well apart from pretty much every econometrics textbook on the market. Angrist and Pischke use real-world econometric examples from the research literature to illustrate each topic in a very applied way. The underlying theory is clearly articulated in the text, as well as explored in more (mathematical) detail in the appendix to each chapter.

The topic coverage is a broad sweep across the main tools in the modern econometrics toolkit: randomised trials, regression models, instrumental variables, regression discontinuity designs, and differences-in-differences, which Angrist and Pischke label the 'Furious Five' of econometric research, with a hat tip to Kung Fu Panda. Like their other book, the key goal the authors have in mind is for budding econometricians to be able to extract causal estimates of relationships from a variety of data.

I think this may be the most enjoyable econometrics book I have ever read. Many of you are probably thinking that statement doesn't set a very high bar, because most econometrics books are drier than a particular arid corner of the Sahara, and almost impenetrable to anyone without high-level mathematics skills. However, there is an astronomical gap between this book and its nearest rivals. If I was teaching undergraduate econometrics, I would not hesitate to use this as the required textbook.

And even if you're not an economics student, but you want to understand the quantitative methods that underlie a large proportion of economics research (including many of the papers that I discuss on this blog), this book would be a great place to start. Highly recommended!