Causality: Models, Reasoning and Inference [Hardcover]

Judea Pearl (Author)

Book Description

Publication Date: September 14, 2009 | ISBN-10: 052189560X | ISBN-13: 978-0521895606 | Edition: 2nd

Written by one of the preeminent researchers in the field, this book provides a comprehensive exposition of modern analysis of causation. It shows how causality has grown from a nebulous concept into a mathematical theory with significant applications in the fields of statistics, artificial intelligence, economics, philosophy, cognitive science, and the health and social sciences. Judea Pearl presents and unifies the probabilistic, manipulative, counterfactual, and structural approaches to causation and devises simple mathematical tools for studying the relationships between causal connections and statistical associations. The book will open the way for including causal analysis in the standard curricula of statistics, artificial intelligence, business, epidemiology, social sciences, and economics. Students in these fields will find natural models, simple inferential procedures, and precise mathematical definitions of causal concepts that traditional texts have evaded or made unduly complicated. The first edition of Causality has led to a paradigmatic change in the way that causality is treated in statistics, philosophy, computer science, social science, and economics. Cited in more than 5,000 scientific publications, it continues to liberate scientists from the traditional molds of statistical thinking. In this revised edition, Judea Pearl elucidates thorny issues, answers readers' questions, and offers a panoramic view of recent advances in this field of research. Causality will be of interests to students and professionals in a wide variety of fields. Anyone who wishes to elucidate meaningful relationships from data, predict effects of actions and policies, assess explanations of reported events, or form theories of causal understanding and causal speech will find this book stimulating and invaluable.

Editorial Reviews

Review

"Make no mistake about it: This is an important book.... The field has no shortage of lively controversy and divergent opinion, but be that as it may, this is certainly one of the contributions that will bring this material further out of the closet and into the face of the broader statistical community, a move that we should welcome both as consumers and as testers of its utility."
Journal of the American Statistical Association


"Pearl's career has been motivated by problems of artificial intelligence, but the implications of this book are much broader. The distinctions he raises and the mathematical foundation he assembles are critical for every field of scientific endeavor. This updated edition of a modern classic deserves a broad and attentive audience."
H. Van Dyke Parunak, Computing Reviews

"Pearl's book is about probabilistic approaches to causality and it gives, especially, empirical researchers working with observational data an immense aid to their research. It also gives theoretical statisticians something to think about as it raises many issues of estimation for example in respective data generating processes. ... This work of Pearl's is an invaluable contribution to the current discussion on the topic of causal modeling. As described by the author his main objective of the book is to develop a framework that integrates substantive knowledge including counterfactuals (through new notations and concepts) with statistical data so as to refine the former and to interpret the latter."
Priyantha Wijayatunga, Significance

Book Description

Written by one of the preeminent researchers in the field, this book provides a comprehensive exposition of modern analysis of causation. It shows how causality has grown from a nebulous concept into a mathematical theory with significant applications in the fields of statistics, artificial intelligence, philosophy, cognitive science, and the health and social sciences.

See all Editorial Reviews

Product Details

• Hardcover: 484 pages
• Publisher: Cambridge University Press; 2nd edition (September 14, 2009)
• Language: English
• ISBN-10: 052189560X
• ISBN-13: 978-0521895606
• Product Dimensions: 8.5 x 1.2 x 10 inches
• Shipping Weight: 2.2 pounds (View shipping rates and policies)
• Average Customer Review: 4.7 out of 5 stars  See all reviews (10 customer reviews)
• Amazon Best Sellers Rank: #29,617 in Books (See Top 100 in Books)
  • #31 in Books > Medical Books > Psychology > Social Psychology & Interactions

More About the Author

Discover books, learn about writers, read author blogs, and more.

Most Helpful Customer Reviews

50 of 51 people found the following review helpful

5.0 out of 5 stars Read this book; here's what you need to know January 30, 2011

By Samuel W. Mitchell

Format:Hardcover|Amazon Verified Purchase

If you are at all capable of understanding it, you must read this book. It gives a general, and theoretical, overview of a highly promising and quite technical theory of what causes are and how to use them in experiments and reasoning. This is applied to practical examples in a very wide range of fields. This is a major step forward in understanding causal reasoning specifically, and scientific reasoning generally.

If you haven't read the first edition:
First, read the Epilogue. Don't start at the beginning. The epilogue will tell you why you should read the book. The book is technical. It is more than worth the effort to follow it.
To follow the mathematics you need a thorough grip on basic probability theory. That is, reasoning using conditional probabilities, conjunctions, independent variables, confounding variables - that sort of thing. You also need the basics of graph theory. You really need to be comfortable with these. The reasoning is very sophisticated, even though the mathematics is basic. It is helpful (but not essential) to know the following too: symbolic logic, basic statistics, some Macroeconomics, some computer science and (occasionally) a little vector algebra.
If you have basic probability and know what a graph is, you ought to read the book.

If you read the first edition:
The second edition repeats the first edition verbatim, but at the end of most chapters there's a clearly defined section dealing with subsequent developments. There's a long chapter at the end that updates you on the replies to the first edition, and some helpful new material explaining things (like d-separation) that were tricky the first time through. Some of this is on the author's website too. The updates are concise. Replies to philosophers (at least) are ultimately devastating, although Pearl could explain himself more fully.
I am a philosopher of science.

34 of 36 people found the following review helpful

5.0 out of 5 stars some interesting questions February 11, 2010

By Steve

Format:Hardcover|Amazon Verified Purchase

In the introductory material, the book claims the graphical method presented in this book 'solves' the problem of causality. However, the book does not read as if the problem has been solved. Instead, it reads like an extended discussion/argument with philosophers, scientists, and statisticians. The book raises a great many interesting questions (some it raises only implicitly), so for this reason I give it 5 stars without hesitation. I do recommend, though, that the third edition of this book substantially reorganize the material; for example, the excellent epilogue should be brought forward as introductory material (and expanded).

27 of 31 people found the following review helpful

5.0 out of 5 stars Causality December 5, 2009

By Patricia Vila Mejia

Format:Hardcover|Amazon Verified Purchase

This is a very suggestive analysis on a quite forgotten by now subject: the study of causality in the social sciences. The author traces very much the original idea of Havelmmo on the nature of econometrics, and brings up to date in the study of several strands of social phenomena that have to do with the nature of causation in human behaviour. He makes use of the notions of bayesian statistics, probability theory, graph theory, correlation analysis and the otherwise called non recursive hierarchical models in social studies. Recommended to those persons who still believe one of the purposes of social studies is to identify and measure causal chains and mechanisms and not simply to focus on correlations and forecasting techniques without due regard to the notion of what causes what and how does it seem to operate in reality.