What do you think?
Rate this book
Mastering 'Metrics: The Path from Cause to Effect
‘Metrics, that is, the field of applied econometrics, encompasses the statistical methods economists use to untangle cause and effect in human affairs. Through accessible discussion and with a dose of kung fu–themed humor, Mastering ’Metrics presents the essential tools of econometric research and demonstrates why econometrics is exciting and useful.
The ‘metrics tools in Mastering ‘Metrics are explained by Joshua D. Angrist (Master Joshway), the Ford Professor of Economics at the Massachusetts Institute of Technology, and Jörn-Steffen Pischke (Master Stevefu), Professor of Economics at the London School of Economics and Political Science.
The ‘metrics tools in Mastering ‘Metrics are explained by Joshua D. Angrist (Master Joshway), the Ford Professor of Economics at the Massachusetts Institute of Technology, and Jörn-Steffen Pischke (Master Stevefu), Professor of Economics at the London School of Economics and Political Science.
293 pages, Kindle Edition
First published December 21, 2014
363 people are currently reading
2039 people want to read
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 30 of 72 reviews
June 19, 2020
Hot take but the fact the whole book is written in kung fu puns added to the experience.
June 4, 2021
This book is a GODSEND. It succeeds in making econometrics, perhaps the driest subject in the world, loads of fun. With an entertaining style (and somewhat corny sense of humor), the book illustrates what the authors deem the "furious five" of econometrics -- randomized trials, regression, instrumental variables, regression discontinuity design, and differences-in-differences -- through a series of fascinating case studies drawn from literature.
There's lots of math to be found in the book, but the authors gently walk the reader through each equation; I'd say little more than a basic knowledge of algebra is required (although some fluency with probability wouldn't hurt). The more gnarly statistical proofs seem to be left to their other more advanced book Mostly Harmless Econometrics. There are also a lot of really helpful and well-made figures and tables to make things clear. I'll probably be keeping this book near my desk at all times -- it really is an invaluable reference. I'm forever thankful that these LEGENDS of economics decided to write a dumb little book for undergraduates.
There's lots of math to be found in the book, but the authors gently walk the reader through each equation; I'd say little more than a basic knowledge of algebra is required (although some fluency with probability wouldn't hurt). The more gnarly statistical proofs seem to be left to their other more advanced book Mostly Harmless Econometrics. There are also a lot of really helpful and well-made figures and tables to make things clear. I'll probably be keeping this book near my desk at all times -- it really is an invaluable reference. I'm forever thankful that these LEGENDS of economics decided to write a dumb little book for undergraduates.
June 28, 2020
Probably a 4.5. A solid and digestible explanation of 3 different econometric approaches. The language was understandable and clear while the examples provided a dynamism to each chapter. I still don’t understand things immediately through formulas, so coupling those with summaries at the end of each chapter was great.
February 14, 2023
Building on my introductory data analysis course, some of this is review, but in the context of bridging the quantitative we learned last quarter and connecting it accurately and thoroughly to the qualitative policy analysis we've been working on this quarter, its utility is essential. Focusing on digging into policy studies, understanding and deconstructing statistical tables, and navigating the precise language to be used when discussing and explaining data, validity, and results to others. I worked off the Kindle edition which was extremely useful for highlights and saving notes, but considering picking up a physical copy to keep around since it's a relatively short book. Recommended for the process of breaking policy studies into digestible pieces and understanding how it all fits together, and how to communicate and discuss with others.
January 3, 2025
Book keeps its promise, but Kung Fu Panda metaphors become a bit annoying at some point.
June 24, 2019
Very good exlanation of the problem of causal inference and ceteris paribus effects in applied economics. For a practical research it has to be supplemented with some standard econometrics textbook or at least with a description of methods in concrete software packages (Stata, R). Inspiring for further research and very useful source for teaching econometrics by real world examples from previous research.
December 30, 2022
Haven’t taken stats since senior year of high school - this read like an in depth yet mildly engaging textbook on practical applications of experimentation and regression analysis in the real world (I only read the sections useful for work)
April 21, 2022
A truly glorious text. When I picked this up I didn't expect that an econometrics book would make me laugh out loud. This book sets itself apart from other economics texts because it contains great explanations of interesting studies in an entertaining way. While the book caters to a rather niche audience, I appreciated the enthusiasm and humor the authors sprinkle throughout the pages to make the material accessible. We desperately need more books like these in economics.
December 11, 2023
5 stars for causal inference!
Honestly a great approach to econometrics with some fun academic humor along the way
Honestly a great approach to econometrics with some fun academic humor along the way
January 9, 2025
Nice recap of basic methods when you took your last metrics curse a while ago.
April 7, 2025
Read for masters program
April 16, 2017
Terribly fun and informative econometrics. Developed an in-depth understanding of various approaches to Linear regression applications. Highly recommended for Stat nerds and math geeks ;)
April 29, 2024
Shut up this was so fun
June 24, 2024
Solid econometrics overview. I would give 3 stars, but I'm not sure how they could have made it more fun and engaging given the topic and broad digestibility, so 4 it is. There are some very good, intuitive explanations.
Quotes/notes
Randomized trials
- "Because E[Yi] is a fixed feature of a particular population, we call it a parameter. Quantities that vary from one sample to another, such as the sample average, are called sample statistics."
- "The standard deviation of a statistic like the sample average is called its standard error.... Every estimate discussed in this book has an associated standard error....This includes sample means ... differences in sample means..., regression coefficients..., and instrumental variables and other more sophisticated methods. Formulas for standard errors can get complicated, but the idea remains simple. The standard error summarizes the variability in an estimate due to random sampling."
- "We might also turn the question of statistical significance on its side: instead of checking whether the sample is consistent with a specific value of mu, we can construct the set of all values of mu that are consistent with the data. The set of such values is called a confidence interval for E[Yi]."
- "Bear in mind that t-statistics and confidence intervals have little to say about whether findings are substantively large or small. A large t-statistic arises when the estimated effect of interest is large but also when the associated standard error is small (as happens when you're blessed with a large sample).... Conversely t-statistics may be small either because the difference in the estimated averages is small or because the standard error of this difference is large."
- "When the path to random assignment is blocked, we look for alternate routes to causal knowledge. Wielded skillfully, [econo]metrics tools other than random assignment can have much of the causality-revealing power of a real experiment. The most basic of these tools is regression, which compares treatment and control subjects who have the same observable characteristics. ... Regression-based causal inference is predicated on the assumption that when key observed variables have been made equal across treatment and control groups, selection bias from the things we can't see is also mostly eliminated."
Regression
- "Alas, there's more to earnings that sex, schools, and SAT scores. Since college attendance decisions aren't randomly assigned, we must control for all factors that determine both attendance decisions and later earnings.... Control for such a wide range of factors seems daunting: the possibilities are virtually infinite and many characteristics are hard to quantify. But Stacy Berg Dale and Alan Krueger came up with a clever and compelling shortcut. Instead of identifying everything that might matter for college choice and earnings, they work with a key summary measure: the characteristics of colleges to which students applied and were admitted." ... "Within each group [ivy, state, etc], students are likely to have similar career ambitions, while they were also judged to be of similar ability by admissions staff at the schools to which they applied. Within-group comparisons should therefore be considerably more apples-to-apples than uncontrollable comparisons involving all students." ... "Evidence of selection bias emerges from a comparison of average earnings across (instead of within) groups A and B."
- "Use of a logged dependent variable allows regression estimates to be interpreted as percent change."
Instrumental Variables (IV)
- "gives IV [instrumental variable] its power, because the instrument changes her treatment status."
- "Can we test the exclusion restrictions? Not directly, but, as is often the case, evidence can be brought to bear on the question."
- "Regression on a constant term and a single dummy variable produces the difference in the conditional means of the dependent variable with the dummy switched off and on."
- 'IV has 3 layers: (1) first stage - requires instruments that affect the causal channel of interest, (2) independence assumption - requires instruments to be as good as randomly assigned, (3) exclusion restriction - a single causal channel connects instruments with outcomes. The 1st be checked by looking for a strong relationship between instruments and the proposed causal channel, the 2nd by checking covariance balance with the intrument switched on and off, the 3rd is not easily verified.'
- "Two-stage least squares (2SLS) generalized IV in two ways. First, 2SLS estimates use multiple instruments efficiently. Second, 2SLS estimates control for covariates, thereby mitigating OVB [ommitted variable bias] from imperfect instruments."
Regression Discontinuity (RD)
- "The parameter.... describes the width of the window and is called a bandwidth.... In practice, bandwidth choice-like the choice of polynomial in parametric models-requires a judgement call. The goal here is not so much to find the one perfect bandwidth as to show that the findings generated by any particular choice of bandwidth are not a fluke."
- "The RD design exploits abrupt changes in treatment status that arise when treatment is determined by a cutoff.... sharp is when treatment itself switches on or off at the cutoff. Fuzzy is when the probability or intensity of treatment jumps. In fuzzy designs, a dummy for clearing the cutoff becomes an instrument: fuzzy design is analyzed by 2SLS."
Differences-in-Differences (DD)
- "Treatment and control groups may differ in the absence of treatment, yet move in parallel. The pattern opens the door to DD estimation of causal effects..... Comparing changes instead of levels, we eliminate fixed differences between groups that might otherwise generate omitted variable bias."
Example: The wages of schooling
- "Good questions are the foundation of our work, and the question of whether increased education really increases earnings is a classic."
- "Variables measured before the treatment variable was determined are generally good controls, because they can't be changed by the treatment."
Quotes/notes
Randomized trials
- "Because E[Yi] is a fixed feature of a particular population, we call it a parameter. Quantities that vary from one sample to another, such as the sample average, are called sample statistics."
- "The standard deviation of a statistic like the sample average is called its standard error.... Every estimate discussed in this book has an associated standard error....This includes sample means ... differences in sample means..., regression coefficients..., and instrumental variables and other more sophisticated methods. Formulas for standard errors can get complicated, but the idea remains simple. The standard error summarizes the variability in an estimate due to random sampling."
- "We might also turn the question of statistical significance on its side: instead of checking whether the sample is consistent with a specific value of mu, we can construct the set of all values of mu that are consistent with the data. The set of such values is called a confidence interval for E[Yi]."
- "Bear in mind that t-statistics and confidence intervals have little to say about whether findings are substantively large or small. A large t-statistic arises when the estimated effect of interest is large but also when the associated standard error is small (as happens when you're blessed with a large sample).... Conversely t-statistics may be small either because the difference in the estimated averages is small or because the standard error of this difference is large."
- "When the path to random assignment is blocked, we look for alternate routes to causal knowledge. Wielded skillfully, [econo]metrics tools other than random assignment can have much of the causality-revealing power of a real experiment. The most basic of these tools is regression, which compares treatment and control subjects who have the same observable characteristics. ... Regression-based causal inference is predicated on the assumption that when key observed variables have been made equal across treatment and control groups, selection bias from the things we can't see is also mostly eliminated."
Regression
- "Alas, there's more to earnings that sex, schools, and SAT scores. Since college attendance decisions aren't randomly assigned, we must control for all factors that determine both attendance decisions and later earnings.... Control for such a wide range of factors seems daunting: the possibilities are virtually infinite and many characteristics are hard to quantify. But Stacy Berg Dale and Alan Krueger came up with a clever and compelling shortcut. Instead of identifying everything that might matter for college choice and earnings, they work with a key summary measure: the characteristics of colleges to which students applied and were admitted." ... "Within each group [ivy, state, etc], students are likely to have similar career ambitions, while they were also judged to be of similar ability by admissions staff at the schools to which they applied. Within-group comparisons should therefore be considerably more apples-to-apples than uncontrollable comparisons involving all students." ... "Evidence of selection bias emerges from a comparison of average earnings across (instead of within) groups A and B."
- "Use of a logged dependent variable allows regression estimates to be interpreted as percent change."
Instrumental Variables (IV)
- "gives IV [instrumental variable] its power, because the instrument changes her treatment status."
- "Can we test the exclusion restrictions? Not directly, but, as is often the case, evidence can be brought to bear on the question."
- "Regression on a constant term and a single dummy variable produces the difference in the conditional means of the dependent variable with the dummy switched off and on."
- 'IV has 3 layers: (1) first stage - requires instruments that affect the causal channel of interest, (2) independence assumption - requires instruments to be as good as randomly assigned, (3) exclusion restriction - a single causal channel connects instruments with outcomes. The 1st be checked by looking for a strong relationship between instruments and the proposed causal channel, the 2nd by checking covariance balance with the intrument switched on and off, the 3rd is not easily verified.'
- "Two-stage least squares (2SLS) generalized IV in two ways. First, 2SLS estimates use multiple instruments efficiently. Second, 2SLS estimates control for covariates, thereby mitigating OVB [ommitted variable bias] from imperfect instruments."
Regression Discontinuity (RD)
- "The parameter.... describes the width of the window and is called a bandwidth.... In practice, bandwidth choice-like the choice of polynomial in parametric models-requires a judgement call. The goal here is not so much to find the one perfect bandwidth as to show that the findings generated by any particular choice of bandwidth are not a fluke."
- "The RD design exploits abrupt changes in treatment status that arise when treatment is determined by a cutoff.... sharp is when treatment itself switches on or off at the cutoff. Fuzzy is when the probability or intensity of treatment jumps. In fuzzy designs, a dummy for clearing the cutoff becomes an instrument: fuzzy design is analyzed by 2SLS."
Differences-in-Differences (DD)
- "Treatment and control groups may differ in the absence of treatment, yet move in parallel. The pattern opens the door to DD estimation of causal effects..... Comparing changes instead of levels, we eliminate fixed differences between groups that might otherwise generate omitted variable bias."
Example: The wages of schooling
- "Good questions are the foundation of our work, and the question of whether increased education really increases earnings is a classic."
- "Variables measured before the treatment variable was determined are generally good controls, because they can't be changed by the treatment."
August 26, 2018
When I build up the courage to tackle an academic and educational book, these are the types of books I look for. Incredibly insightful but also wonderfully personable. I feel like I'm having a conversation with a very knowledgeable fellow rather than being lectured and constantly reminded that I have a long way to go before I am an expert.
There are of course parts of this book that you will find dull. Consider noting those parts because that might signify your lack of personal interest in that area. Also, don't feel pressured to understand every little detail mentioned. Some are heavier than others. Lastly, don't be afraid of purchasing this book to keep as a resource. I've already used it a few times since reading.
There are of course parts of this book that you will find dull. Consider noting those parts because that might signify your lack of personal interest in that area. Also, don't feel pressured to understand every little detail mentioned. Some are heavier than others. Lastly, don't be afraid of purchasing this book to keep as a resource. I've already used it a few times since reading.
November 15, 2016
I'm a fan of this method - take a few key points and explain them simply. I really like how Angrist and Pischke break down and discuss the topics covered (randomized trials, regression, instrumental variables, regression discontinuity designs, differences-in-differences). This book should be on the shelf of anyone interested in doing applied policy work who aren't math geniuses.
October 1, 2015
Insightful and helpful in developing intuition for theorizing and building models and addressing selection bias. The Kung Fu references were just a bit irritating.
March 23, 2019
First, the content. Mastering 'Metrics does a pretty good job of covering the intuition (and some of the math) behind random assignment, regression, instrumental variables, regression discontinuity designs, and difference in differences. I think their treatment of these topics would be most useful to someone who was trying to read modern applied econometrics (or political science). Ideally the reader would have taken enough statistics that they can focus on trying to grasp the concept of potential outcomes rather than trying to work through the algebra. The methods that are covered are extremely important in social science and so having an idea of what they do and why we use them is helpful.
It leaves a lot to be desired, though. I'm honestly not sure who they think their primary audience is. The content is aimed primarily at an undergraduate level. Great! Undergrads deserve to be exposed to causal inference! But Mastering 'Metrics doesn't actually give you any of the tools you would need to *do* causal inference. There are a couple of vague references to statistical software (Stata) but the idea that a student would read this and then go off to perform DiD is laughable. There is nothing wrong with having an accessible primer on the intuition behind an idea, but if you're doing that you should probably be upfront about it and make recommendations on where to go from here. I suspect that the assumption is that the reader will next go pick up their other book "Mostly Harmless Econometrics" but that is never explicitly stated.
Personally I found the extended metaphor that econometrics is kung fu to be annoying. I think the authors believed that they were making the material more accessible by treating it less reverently, which I agree could have been an effective communication strategy, but I think it mostly fell flat. If I'm cringing at your puns I'm not learning about local average treatment effects. Moreover, I think the metaphor that econometrics is kung fu is actually harmful. Kung fu is mysterious and mystical. It's studied at the feet of a master over the course of a lifetime. The master might have you wash floors for a year, without offering a reason. There is definitely an art to econometrics, but clouding econometrics in mysticism does more to protect the reputation of the teacher than it does to advance the student's learning. Others may disagree but this grasshopper would have preferred we spend less time in the dojo and more time in the computer lab.
Which brings me to my final point. I believe that this book was written by economists who see data science expanding into areas that they consider to be economics and are uncomfortable with this development. The jacket blurb calls econometrics the "original data science" and argues that it's "exciting and useful." The people they asked to provide blurbs for the back include Hal Varian (of Google), and Andrew Gelman (of Bayesian and Stan fame). There's nothing wrong with aiming econometrics at data scientists, in fact I think there is a lot that they can and should learn from each other. The problem is that neither sees the other as particularly useful or intelligent and efforts at communication inevitably betray this bias.
Economists view data scientists as regression monkeys (probably the worst insult you can give someone in economics). When they look at data science they just see extremely elaborate efforts at curve fitting. Since economists don't think curve fitting is all that interesting or useful for doing economics, they scoff at neural networks and boosting. Imagine their horror when they see data science moving into their territory.
Data scientists, on the other hand, don't often think about economics at all. From their perspective the two disciplines have basically no overlap. So they struggle to see why they should care about what an economist has to say about anything. This is primarily driven by the popular misperception of economics being about business questions. Imagine their frustration when economists start telling them that their results are wrong.
The fact is that the two do have a lot to share with each other. But I've wasted enough of your time on this tangent so I'll save that for another time.
It leaves a lot to be desired, though. I'm honestly not sure who they think their primary audience is. The content is aimed primarily at an undergraduate level. Great! Undergrads deserve to be exposed to causal inference! But Mastering 'Metrics doesn't actually give you any of the tools you would need to *do* causal inference. There are a couple of vague references to statistical software (Stata) but the idea that a student would read this and then go off to perform DiD is laughable. There is nothing wrong with having an accessible primer on the intuition behind an idea, but if you're doing that you should probably be upfront about it and make recommendations on where to go from here. I suspect that the assumption is that the reader will next go pick up their other book "Mostly Harmless Econometrics" but that is never explicitly stated.
Personally I found the extended metaphor that econometrics is kung fu to be annoying. I think the authors believed that they were making the material more accessible by treating it less reverently, which I agree could have been an effective communication strategy, but I think it mostly fell flat. If I'm cringing at your puns I'm not learning about local average treatment effects. Moreover, I think the metaphor that econometrics is kung fu is actually harmful. Kung fu is mysterious and mystical. It's studied at the feet of a master over the course of a lifetime. The master might have you wash floors for a year, without offering a reason. There is definitely an art to econometrics, but clouding econometrics in mysticism does more to protect the reputation of the teacher than it does to advance the student's learning. Others may disagree but this grasshopper would have preferred we spend less time in the dojo and more time in the computer lab.
Which brings me to my final point. I believe that this book was written by economists who see data science expanding into areas that they consider to be economics and are uncomfortable with this development. The jacket blurb calls econometrics the "original data science" and argues that it's "exciting and useful." The people they asked to provide blurbs for the back include Hal Varian (of Google), and Andrew Gelman (of Bayesian and Stan fame). There's nothing wrong with aiming econometrics at data scientists, in fact I think there is a lot that they can and should learn from each other. The problem is that neither sees the other as particularly useful or intelligent and efforts at communication inevitably betray this bias.
Economists view data scientists as regression monkeys (probably the worst insult you can give someone in economics). When they look at data science they just see extremely elaborate efforts at curve fitting. Since economists don't think curve fitting is all that interesting or useful for doing economics, they scoff at neural networks and boosting. Imagine their horror when they see data science moving into their territory.
Data scientists, on the other hand, don't often think about economics at all. From their perspective the two disciplines have basically no overlap. So they struggle to see why they should care about what an economist has to say about anything. This is primarily driven by the popular misperception of economics being about business questions. Imagine their frustration when economists start telling them that their results are wrong.
The fact is that the two do have a lot to share with each other. But I've wasted enough of your time on this tangent so I'll save that for another time.
May 15, 2024
This book provides a great informal and practical overview of the strategies used to infer causality in modern economics research. It explains these strategies by breaking down landmark studies that use them. It also presents a clear picture of the statistical methods without delving deep into the maths. Overall, I found it a great complement to a more formal econometrics class. It is certainly not technical enough to teach you how to use the methods, but it is still a good and accessible starting point.
The book goes over:
Randomised Control Trials: e.g. medical trials - the researcher constructs two similar groups and assigns a treatment to one of them to assess its effect. Unfortunately, it is almost impossible to conduct in most economic contexts.
Regression: This technique controls for possible confounders—e.g., more educated people earn more. However, children of richer parents are also more educated and earn more. To separate the effect of education on wages, we use regression to find the effect of education on two people with the same parental income. This is a fairly rudimentary technique that is used in every empirical research paper.
Instrumental variable: Imitating a randomised trial by finding a naturally occurring process that is similar to random assignment - e.g. Charter school attendees earn more than public school attendees. But, people who apply to charter schools are more motivated and might have been successful anyway. in some US states, overfilled schools choose applicants via lottery. This means that the only difference between successful and unsuccessful applicants is their luck in the draw, not their motivation. Therefore, the differences in earnings between the successful and unsuccessful applicants are only because of their schooling.
Regression discontinuity: Using arbitrary breaks to study causal effects—e.g., the only difference between people who are 17.95 years old and those who are 18.05 years old is that they can legally drink. This means that the difference between the two age groups should only reflect the effect of legal alcohol drinking.
Differences in differences: Comparing the treated group with the general trend - e.g. Israel liberalised after a wave of Russian Jew immigrants. But the whole region liberalised - was it the Russians or the trend? We look at the relative standing in the region - perhaps it was the 10. most liberalised country before, and now, it is the 2. most liberalised. This should reflect just the event we're studying.
The book goes over:
Randomised Control Trials: e.g. medical trials - the researcher constructs two similar groups and assigns a treatment to one of them to assess its effect. Unfortunately, it is almost impossible to conduct in most economic contexts.
Regression: This technique controls for possible confounders—e.g., more educated people earn more. However, children of richer parents are also more educated and earn more. To separate the effect of education on wages, we use regression to find the effect of education on two people with the same parental income. This is a fairly rudimentary technique that is used in every empirical research paper.
Instrumental variable: Imitating a randomised trial by finding a naturally occurring process that is similar to random assignment - e.g. Charter school attendees earn more than public school attendees. But, people who apply to charter schools are more motivated and might have been successful anyway. in some US states, overfilled schools choose applicants via lottery. This means that the only difference between successful and unsuccessful applicants is their luck in the draw, not their motivation. Therefore, the differences in earnings between the successful and unsuccessful applicants are only because of their schooling.
Regression discontinuity: Using arbitrary breaks to study causal effects—e.g., the only difference between people who are 17.95 years old and those who are 18.05 years old is that they can legally drink. This means that the difference between the two age groups should only reflect the effect of legal alcohol drinking.
Differences in differences: Comparing the treated group with the general trend - e.g. Israel liberalised after a wave of Russian Jew immigrants. But the whole region liberalised - was it the Russians or the trend? We look at the relative standing in the region - perhaps it was the 10. most liberalised country before, and now, it is the 2. most liberalised. This should reflect just the event we're studying.
December 3, 2021
Alternate title: "Causal inference for the ambitious beginner".
While very readable and conversational in tone, it is still rather terse in some sections as if you're expected to already have seen most of the material here on a previous introductory course. I'd say you should look elsewhere if the finer points of regression and OLS are not fresh on your mind.
The material covered on causal inference is as follows: randomized trials, regression, instrumental variables, regression discontinuity design, and differences-in-differences.
Of the above, I was only really familiar with the first two from undergrad. The presentation and exposition on the remaining three topics are very good but the authors seem to gloss over some detail for the sake of brevity which will make the text less than ideal for true beginners in econometrics.
Still, plenty useful and a reference to further excursions into the subject. Recommended!
While very readable and conversational in tone, it is still rather terse in some sections as if you're expected to already have seen most of the material here on a previous introductory course. I'd say you should look elsewhere if the finer points of regression and OLS are not fresh on your mind.
The material covered on causal inference is as follows: randomized trials, regression, instrumental variables, regression discontinuity design, and differences-in-differences.
Of the above, I was only really familiar with the first two from undergrad. The presentation and exposition on the remaining three topics are very good but the authors seem to gloss over some detail for the sake of brevity which will make the text less than ideal for true beginners in econometrics.
Still, plenty useful and a reference to further excursions into the subject. Recommended!
June 8, 2023
Excelente.
Metric’s é como os atuantes na área chamam carinhosamente o campo de Econometria (Econometrics).
A grande maestria aqui é a forma de explicar. Os autores possuem uma didática excelente e conseguem inserir humor e leveza em assunto tao complexo. Os autores criaram personagens inspirados na série norte americana Kung Fu.
Não é um livro para iniciantes no assunto em causalidade, e recomendo que você tenha alguma noção de estatística básica. O livro é bem curto e aborda as principais ferramentas de econometria para inferência causal. Não há codigos aqui, apenas teoria com exemplos de dados e experimentos reais.
Um dos autores, Joshua (Nobel) possui um canal no youtube com o mesmo assunto e explica bem as ferramentas de inferência causal de uma forma bem lúdica. Veja alguns vídeos antes de adquirir o livro :)
Recomendo para quem atua com ciência de dados de uma forma geral.
Metric’s é como os atuantes na área chamam carinhosamente o campo de Econometria (Econometrics).
A grande maestria aqui é a forma de explicar. Os autores possuem uma didática excelente e conseguem inserir humor e leveza em assunto tao complexo. Os autores criaram personagens inspirados na série norte americana Kung Fu.
Não é um livro para iniciantes no assunto em causalidade, e recomendo que você tenha alguma noção de estatística básica. O livro é bem curto e aborda as principais ferramentas de econometria para inferência causal. Não há codigos aqui, apenas teoria com exemplos de dados e experimentos reais.
Um dos autores, Joshua (Nobel) possui um canal no youtube com o mesmo assunto e explica bem as ferramentas de inferência causal de uma forma bem lúdica. Veja alguns vídeos antes de adquirir o livro :)
Recomendo para quem atua com ciência de dados de uma forma geral.
December 1, 2021
A solid grounding in core econometrics techniques. I do hope that most of my time in my data career I can jump straight to the gold standard of randomized controlled tests, but sometimes that's not feasible, and this book offers a wealth of alternative solutions. I also appreciate the asides that provide historical context on the past masters who developed these techniques. I did occasionally find myself raising an eyebrow at nuances in the data that went unexplored, and there were a few explanations that could have been stated more clearly (thinking here of a section in chapter 6 that should have been nearly summed up as "don't violate the laws of (temporal) causality in your regression").
November 14, 2024
I have to say that this book is helpful but can get complicated quickly. I wish that I had started this book when I was taking my Econometrics class last year, and read this alongside it, because it would have helped me. If you are trying to review Econometrics like what I was doing, then this book would not be as helpful for you. I can appreciate the approach the authors tried to make in their explanations making it fun and having good examples, but it was still complicated for me. This book is one of those books that I will probably reference a lot in the future, alongside other econometric sources.
December 18, 2024
I loved this book so much. I had to read this book for an Econometrics class, and I felt it was the perfect introduction to key econometric concepts, such as regression, difference in difference, regression discontinuity, and instrument variables. The key to this book is that it help explain the mathematical ideas in a very clear conceptual manner so that you understand not only how the techniques work, but also when you might apply them and why. They do all this while also maintaining a humor that kept me engaged throughout. For anyone looking to gain a solid foundational non-mathematical understanding of some key econometrics ideas, I would definitely recommend this book.
May 22, 2023
Good introductory book for an undergrad or early master's student. While it adequately covers core econometric concepts, I'm not convinced it effectively balances a textbook approach vs that of a "how-to" guide: it likely would have been better suited leaning a bit more in one direction or another.
Definitely fairly dense, and would best be accompanied by some examples you can work through. Wonderful companion to anyone working through an introductory Python or R course, as the technical implementations would complement this book's explanations well.
Definitely fairly dense, and would best be accompanied by some examples you can work through. Wonderful companion to anyone working through an introductory Python or R course, as the technical implementations would complement this book's explanations well.
July 9, 2016
The positives of this book are instantly revealed to those who are working on this topic, so for them I am not going to comment much. But to those who want to understand what most economists do these days and what are their methods - I think this book is a neat introduction.
It gives you circumstances where economists thrive with their data work and where they build fascinating analysis from interesting natural accidents. Have you ever imagined that Fred and George (the famous identical twins in Harry Potter) can be interesting test subjects to examine errors on returns to schooling estimates. Twins are used to examine schooling outcomes as they help keep ability fixed - identical twins are said to be similarly abled.
Or have you wondered why we have to measure weird things (data on quarter of births) to understand the impact of education. These and many other issues which are explored in this book actually bring out a glamorous aspect of the toils economists go through in examining an issue with the precision, care and concern - especially because policies are a result of these studies! It is thus an intersting starting place for beginners too! However, my expectations from this book were more - especially since I like the papers written by Angrist etc.
I started reading this book as a 'forced formula' in a standard Labor Economics course but never ended up reading the whole thing, until I decided to pick it up again. I did not expect the details of an econometrics course; but I thought as the book started out, it would have a funny take and make things interesting for the reader. Perhaps even reach out to non-economics students who just would want to venture out closer to the 'metrics way of research.
But I have to accept it! Economists are pretty bad at jokes and they aren't even slapstick funny! There is this disconnect with the jokes and the Kung Fu experience which is tried in the book. Though personally I like the Kung Fu Panda and I feel the idea was interesting and even well placed - its implementation did not flow through.
The chapters I feel are also imbalanced. Take for instance - Chapters on Regression, RDD are flowing smoothly, but the chapter on IV is tighter than the others. On the merit of how much does the book intend to give the reader the details on these things is another issue. But given a cursory exposition on this, I think IV overdoes it, whereas other chapters are more pointed and do not bring out unnecessary details.
There is also an effort at comparison of various techniques and lingering of the IV-2SLS; but I feel either the comparison should have flowed through the entire book, or should have been chapterized separately. In places where the story of a DD is flowing, an IV comparison takes one off guard in terms of now being able to apply and compare.
In terms of the chapters itself, I think they are very topical and will cover a lot of the modern research; the book pulls away from a fundamental issue - no matter what the methods are, the thought of comparison and counterfactuals is not emphasized enough I feel. Consider a standard econometrics textbook - say Wooldridge - it actually draws a framework where you know - no matter what the empirical problem is, you need to think in terms of identification, endogeneity and the underlying logic of counter-factuals. They certainly bring in a lot of that - where they talk about apples-to-apples comparison; but the emphasis is not approached as a general method of empirical analysis and the book can go far if that is emphasized. Thus in terms of binding the various methods - (a) a comparison and (b) a generalized empirical strategy might help get the econometrics logic through to a wider audience.
Another relevant factor with the book is that the passages do not lead you to read on - rather they are too academic! If the intention is for a wider audience and for a more diversified crowd, then the importance of leading readers onto the next issues is of supreme importance. For eg: they are discussing an issue and then the next issue comes up as a next section. There is no sense of direction as to why am I reading about an issue and where do the connections matter - in terms of comprehending the entire topic, the reader is left on his own.
Admitting that the academic way keeps the writing clean, but then it also makes the reader lose interest. The snippets are like the buzz generators - they are the interest makers - and this book could have gone a long long way in making 'Metrics fun!.
[Some notes while I was reading this book -
"So i have almost reached halfway chapter 4 where RDD is being discussed. I found the chapters imbalanced. Like the IV chapter was very heavy and was not a smoother flow like the other ones.
But the IV chapter was better in terms of the details whereas RDD chapter isn't as heavy on those details. So the detailing level has to be consistent. Further there is a need to link the discussions. Suddenly a topic is completed and another section starts with a new topic. This to me seems disconnected and you don't really get the flow in the argument while reading the book.
We don't want a book which gives us examples and then loses us in these examples. For examples should lead us to building of the concepts and continue our quest forward. "
]
It gives you circumstances where economists thrive with their data work and where they build fascinating analysis from interesting natural accidents. Have you ever imagined that Fred and George (the famous identical twins in Harry Potter) can be interesting test subjects to examine errors on returns to schooling estimates. Twins are used to examine schooling outcomes as they help keep ability fixed - identical twins are said to be similarly abled.
Or have you wondered why we have to measure weird things (data on quarter of births) to understand the impact of education. These and many other issues which are explored in this book actually bring out a glamorous aspect of the toils economists go through in examining an issue with the precision, care and concern - especially because policies are a result of these studies! It is thus an intersting starting place for beginners too! However, my expectations from this book were more - especially since I like the papers written by Angrist etc.
I started reading this book as a 'forced formula' in a standard Labor Economics course but never ended up reading the whole thing, until I decided to pick it up again. I did not expect the details of an econometrics course; but I thought as the book started out, it would have a funny take and make things interesting for the reader. Perhaps even reach out to non-economics students who just would want to venture out closer to the 'metrics way of research.
But I have to accept it! Economists are pretty bad at jokes and they aren't even slapstick funny! There is this disconnect with the jokes and the Kung Fu experience which is tried in the book. Though personally I like the Kung Fu Panda and I feel the idea was interesting and even well placed - its implementation did not flow through.
The chapters I feel are also imbalanced. Take for instance - Chapters on Regression, RDD are flowing smoothly, but the chapter on IV is tighter than the others. On the merit of how much does the book intend to give the reader the details on these things is another issue. But given a cursory exposition on this, I think IV overdoes it, whereas other chapters are more pointed and do not bring out unnecessary details.
There is also an effort at comparison of various techniques and lingering of the IV-2SLS; but I feel either the comparison should have flowed through the entire book, or should have been chapterized separately. In places where the story of a DD is flowing, an IV comparison takes one off guard in terms of now being able to apply and compare.
In terms of the chapters itself, I think they are very topical and will cover a lot of the modern research; the book pulls away from a fundamental issue - no matter what the methods are, the thought of comparison and counterfactuals is not emphasized enough I feel. Consider a standard econometrics textbook - say Wooldridge - it actually draws a framework where you know - no matter what the empirical problem is, you need to think in terms of identification, endogeneity and the underlying logic of counter-factuals. They certainly bring in a lot of that - where they talk about apples-to-apples comparison; but the emphasis is not approached as a general method of empirical analysis and the book can go far if that is emphasized. Thus in terms of binding the various methods - (a) a comparison and (b) a generalized empirical strategy might help get the econometrics logic through to a wider audience.
Another relevant factor with the book is that the passages do not lead you to read on - rather they are too academic! If the intention is for a wider audience and for a more diversified crowd, then the importance of leading readers onto the next issues is of supreme importance. For eg: they are discussing an issue and then the next issue comes up as a next section. There is no sense of direction as to why am I reading about an issue and where do the connections matter - in terms of comprehending the entire topic, the reader is left on his own.
Admitting that the academic way keeps the writing clean, but then it also makes the reader lose interest. The snippets are like the buzz generators - they are the interest makers - and this book could have gone a long long way in making 'Metrics fun!.
[Some notes while I was reading this book -
"So i have almost reached halfway chapter 4 where RDD is being discussed. I found the chapters imbalanced. Like the IV chapter was very heavy and was not a smoother flow like the other ones.
But the IV chapter was better in terms of the details whereas RDD chapter isn't as heavy on those details. So the detailing level has to be consistent. Further there is a need to link the discussions. Suddenly a topic is completed and another section starts with a new topic. This to me seems disconnected and you don't really get the flow in the argument while reading the book.
We don't want a book which gives us examples and then loses us in these examples. For examples should lead us to building of the concepts and continue our quest forward. "
]
June 20, 2020
This playful examination of core econometric concepts is an accessible and valuable tool for beginning econometricians. Though at times a touch dry, the authors strive to keep readers engaged. Explanations are clear, concise, and well-supported with examples, derivations, and references. A worthy introductory read.
October 26, 2020
Ohh how you wish more econometrics text to be like this. The book is concise and lucid and helps to apply some of these concepts into the work immediately. It is great that the author separate mathematical derivation and concept intuition, it really helps in understanding the concept. Hope to see many more books from him and kung fu masters (Mostly harmless econometrics already on the list)
September 2, 2021
Simple examples for understanding tools. Since reading it I have been using sentences as: Is your ceteris really paribus? On average, I think that... and similar. Quite funny how it helped me realize how misguided most statements tend to be. Really recommend it to everyone, even people not interested in econometrics.
February 7, 2023
Fun book! (As fun as econometrics can be when you know little to nothing about it)
Ignore this unless you want to learn how to understand statistics, economics, research/survey design, and generally how to read data that people refer to when they pontificate while making vague references to "research."
Ignore this unless you want to learn how to understand statistics, economics, research/survey design, and generally how to read data that people refer to when they pontificate while making vague references to "research."
Displaying 1 - 30 of 72 reviews