Ebook: Causation, Prediction, and Search

This book is intended for anyone, regardless of discipline, who is interested in the use of statistical methods to help obtain scientific explanations or to predict the outcomes of actions, experiments or policies. Much of G. Udny Yule's work illustrates a vision of statistics whose goal is to investigate when and how causal influences may be reliably inferred, and their comparative strengths estimated, from statistical samples. Yule's enterprise has been largely replaced by Ronald Fisher's conception, in which there is a fundamental cleavage between experimental and non­ experimental inquiry, and statistics is largely unable to aid in causal inference without randomized experimental trials. Every now and then members of the statistical community express misgivings about this turn of events, and, in our view, rightly so. Our work represents a return to something like Yule's conception of the enterprise of theoretical statistics and its potential practical benefits. If intellectual history in the 20th century had gone otherwise, there might have been a discipline to which our work belongs. As it happens, there is not. We develop material that belongs to statistics, to computer science, and to philosophy; the combination may not be entirely satisfactory for specialists in any of these subjects. We hope it is nonetheless satisfactory for its purpose.

---

This thoroughly thought-provoking book is unorthodox in its claim that under appropriate assumptions causal structures may be inferred from non-experimental sample data. The authors adopt two axioms relating causal relationships to probability distributions. These axioms have only been explicitly suggested in the statistical literature over the last 15 years but have been implicitly assumed in a variety of statistical disciplines. On the basis of these axioms, the authors propose a number of computationally efficient search procedures that infer causal relationships from non-experimental sample data and background knowledge. They also deduce a variety of theorems concerning estimation, sampling, latent variable existence and structure, regression, indistinguishability relations, experimental design, prediction, Simpsons paradox, and other topics. For the most part, technical details have been placed in the book's last chapter, and so the main results will be accessible to any research worker (regardless of discipline) who is interested in statistical methods to help establish or refute causal claims.

---

This thoroughly thought-provoking book is unorthodox in its claim that under appropriate assumptions causal structures may be inferred from non-experimental sample data. The authors adopt two axioms relating causal relationships to probability distributions. These axioms have only been explicitly suggested in the statistical literature over the last 15 years but have been implicitly assumed in a variety of statistical disciplines. On the basis of these axioms, the authors propose a number of computationally efficient search procedures that infer causal relationships from non-experimental sample data and background knowledge. They also deduce a variety of theorems concerning estimation, sampling, latent variable existence and structure, regression, indistinguishability relations, experimental design, prediction, Simpsons paradox, and other topics. For the most part, technical details have been placed in the book's last chapter, and so the main results will be accessible to any research worker (regardless of discipline) who is interested in statistical methods to help establish or refute causal claims.
Content:
Front Matter....Pages i-xxiii
Introduction and Advertisement....Pages 1-24
Formal Preliminaries....Pages 25-40
Causation and Prediction: Axioms and Explications....Pages 41-86
Statistical Indistinguishability....Pages 87-102
Discovery Algorithms for Causally Sufficient Structures....Pages 103-162
Discovery Algorithms without Causal Sufficiency....Pages 163-200
Prediction....Pages 201-237
Regression, Causation and Prediction....Pages 238-258
The Design of Empirical Studies....Pages 259-305
The Structure of the Unobserved....Pages 306-322
Elaborating Linear Theories with Unmeasured Variables....Pages 323-353
Open Problems....Pages 354-366
Proofs of Theorems....Pages 367-480
Back Matter....Pages 481-529

---

This thoroughly thought-provoking book is unorthodox in its claim that under appropriate assumptions causal structures may be inferred from non-experimental sample data. The authors adopt two axioms relating causal relationships to probability distributions. These axioms have only been explicitly suggested in the statistical literature over the last 15 years but have been implicitly assumed in a variety of statistical disciplines. On the basis of these axioms, the authors propose a number of computationally efficient search procedures that infer causal relationships from non-experimental sample data and background knowledge. They also deduce a variety of theorems concerning estimation, sampling, latent variable existence and structure, regression, indistinguishability relations, experimental design, prediction, Simpsons paradox, and other topics. For the most part, technical details have been placed in the book's last chapter, and so the main results will be accessible to any research worker (regardless of discipline) who is interested in statistical methods to help establish or refute causal claims.
Content:
Front Matter....Pages i-xxiii
Introduction and Advertisement....Pages 1-24
Formal Preliminaries....Pages 25-40
Causation and Prediction: Axioms and Explications....Pages 41-86
Statistical Indistinguishability....Pages 87-102
Discovery Algorithms for Causally Sufficient Structures....Pages 103-162
Discovery Algorithms without Causal Sufficiency....Pages 163-200
Prediction....Pages 201-237
Regression, Causation and Prediction....Pages 238-258
The Design of Empirical Studies....Pages 259-305
The Structure of the Unobserved....Pages 306-322
Elaborating Linear Theories with Unmeasured Variables....Pages 323-353
Open Problems....Pages 354-366
Proofs of Theorems....Pages 367-480
Back Matter....Pages 481-529
....