Record:   Prev Next
Author Berzuini, Carlo
Title Causality : Statistical Perspectives and Applications
Imprint New York : John Wiley & Sons, Incorporated, 2012
©2012
book jacket
Edition 1st ed
Descript 1 online resource (415 pages)
text txt rdacontent
computer c rdamedia
online resource cr rdacarrier
Series Wiley Series in Probability and Statistics Ser
Wiley Series in Probability and Statistics Ser
Note Intro -- Causality: Statistical Perspectives and Applications -- List of contributors -- An overview of statistical causality -- 1 Statistical causality: Some historical remarks -- 1.1 Introduction -- 1.2 Key issues -- 1.3 Rothamsted view -- 1.4 An earlier controversy and its implications -- 1.5 Three versions of causality -- 1.6 Conclusion -- References -- 2 The language of potential outcomes -- 2.1 Introduction -- 2.2 Definition of causal effects through potential outcomes -- 2.2.1 Subject-specific causal effects -- 2.2.2 Population causal effects -- 2.2.3 Association versus causation -- 2.3 Identification of population causal effects -- 2.3.1 Randomized experiments -- 2.3.2 Observational studies -- 2.4 Discussion -- References -- 3 Structural equations, graphs and interventions -- 3.1 Introduction -- 3.2 Structural equations, graphs, and interventions -- 3.2.1 Graph terminology -- 3.2.2 Markovian models -- 3.2.3 Latent projections and semi-Markovian models -- 3.2.4 Interventions in semi-Markovian models -- 3.2.5 Counterfactual distributions in NPSEMs -- 3.2.6 Causal diagrams and counterfactual independence -- 3.2.7 Relation to potential outcomes -- References -- 4 The decision-theoretic approach to causal inference -- 4.1 Introduction -- 4.2 Decision theory and causality -- 4.2.1 A simple decision problem -- 4.2.2 Causal inference -- 4.3 No confounding -- 4.4 Confounding -- 4.4.1 Unconfounding -- 4.4.2 Nonconfounding -- 4.4.3 Back-door formula -- 4.5 Propensity analysis -- 4.6 Instrumental variable -- 4.6.1 Linear model -- 4.6.2 Binary variables -- 4.7 Effect of treatment of the treated -- 4.8 Connections and contrasts -- 4.8.1 Potential responses -- 4.8.2 Causal graphs -- 4.9 Postscript -- Acknowledgements -- References -- 5 Causal inference as a prediction problem: Assumptions, identification and evidence synthesis -- 5.1 Introduction
5.2 A brief commentary on developments since 1970 -- 5.2.1 Potential outcomes and missing data -- 5.2.2 The prognostic view -- 5.3 Ambiguities of observational extensions -- 5.4 Causal diagrams and structural equations -- 5.5 Compelling versus plausible assumptions, models and inferences -- 5.6 Nonidentification and the curse of dimensionality -- 5.7 Identification in practice -- 5.8 Identification and bounded rationality -- 5.9 Conclusion -- Acknowledgments -- References -- 6 Graph-based criteria of identifiability of causal questions -- 6.1 Introduction -- 6.2 Interventions from observations -- 6.3 The back-door criterion, conditional ignorability, and covariate adjustment -- 6.4 The front-door criterion -- 6.5 Do-calculus -- 6.6 General identification -- 6.7 Dormant independences and post-truncation constraints -- References -- 7 Causal inference from observational data: A Bayesian predictive approach -- 7.1 Background -- 7.2 A model prototype -- 7.3 Extension to sequential regimes -- 7.4 Providing a causal interpretation: Predictive inference from data -- 7.5 Discussion -- Acknowledgement -- References -- 8 Assessing dynamic treatment strategies -- 8.1 Introduction -- 8.2 Motivating example -- 8.3 Descriptive versus causal inference -- 8.4 Notation and problem definition -- 8.5 HIV example continued -- 8.6 Latent variables -- 8.7 Conditions for sequential plan identifiability -- 8.7.1 Stability -- 8.7.2 Positivity -- 8.8 Graphical representations of dynamic plans -- 8.9 Abdominal aortic aneurysm surveillance -- 8.10 Statistical inference and computation -- 8.11 Transparent actions -- 8.12 Refinements -- 8.13 Discussion -- Acknowledgements -- References -- 9 Causal effects and natural laws: Towards a conceptualization of causal counterfactuals for nonmanipulable exposures, with application to the effects of race and sex -- 9.1 Introduction
9.2 Laws of nature and contrary to fact statements -- 9.3 Association and causation in the social and biomedical sciences -- 9.4 Manipulation and counterfactuals -- 9.5 Natural laws and causal effects -- 9.6 Consequences of randomization -- 9.7 On the causal effects of sex and race -- 9.8 Discussion -- Acknowledgements -- References -- 10 Cross-classifications by joint potential outcomes -- 10.1 Introduction -- 10.2 Bounds for the causal treatment effect in randomized trials with imperfect compliance -- 10.3 Identifying the complier causal effect in randomized trials with imperfect compliance -- 10.4 Defining the appropriate causal effect in studies suffering from truncation by death -- 10.5 Discussion -- References -- 11 Estimation of direct and indirect effects -- 11.1 Introduction -- 11.2 Identification of the direct and indirect effect -- 11.2.1 Definitions -- 11.2.2 Identification -- 11.3 Estimation of controlled direct effects -- 11.3.1 G-computation -- 11.3.2 Inverse probability of treatment weighting -- 11.3.3 G-estimation for additive and multiplicative models -- 11.3.4 G-estimation for logistic models -- 11.3.5 Case-control studies -- 11.3.6 G-estimation for additive hazard models -- 11.4 Estimation of natural direct and indirect effects -- 11.5 Discussion -- Acknowledgements -- References -- 12 The mediation formula: A guide to the assessment of causal pathways in nonlinear models -- 12.1 Mediation: Direct and indirect effects -- 12.1.1 Direct versus total effects -- 12.1.2 Controlled direct effects -- 12.1.3 Natural direct effects -- 12.1.4 Indirect effects -- 12.1.5 Effect decomposition -- 12.2 The mediation formula: A simple solution to a thorny problem -- 12.2.1 Mediation in nonparametric models -- 12.2.2 Mediation effects in linear, logistic, and probit models -- 12.2.3 Special cases of mediation models -- 12.2.4 Numerical example
12.3 Relation to other methods -- 12.3.1 Methods based on differences and products -- 12.3.2 Relation to the principal-strata direct effect -- 12.4 Conclusions -- Acknowledgments -- References -- 13 The sufficient cause framework in statistics, philosophy and the biomedical and social sciences -- 13.1 Introduction -- 13.2 The sufficient cause framework in philosophy -- 13.3 The sufficient cause framework in epidemiology and biomedicine -- 13.4 The sufficient cause framework in statistics -- 13.5 The sufficient cause framework in the social sciences -- 13.6 Other notions of sufficiency and necessity in causal inference -- 13.7 Conclusion -- Acknowledgements -- References -- 14 Analysis of interaction for identifying causal mechanisms -- 14.1 Introduction -- 14.2 What is a mechanism? -- 14.3 Statistical versus mechanistic interaction -- 14.4 Illustrative example -- 14.5 Mechanistic interaction defined -- 14.6 Epistasis -- 14.7 Excess risk and superadditivity -- 14.8 Conditions under which excess risk and superadditivity indicate the presence of mechanistic interaction -- 14.9 Collapsibility -- 14.10 Back to the illustrative study -- 14.11 Alternative approaches -- 14.12 Discussion -- Ethics statement -- Financial disclosure -- References -- 15 Ion channels as a possible mechanism of neurodegeneration in multiple sclerosis -- 15.1 Introduction -- 15.2 Background -- 15.3 The scientific hypothesis -- 15.4 Data -- 15.5 A simple preliminary analysis -- 15.6 Testing for qualitative interaction -- 15.7 Discussion -- Acknowledgments -- References -- 16 Supplementary variables for causal estimation -- 16.1 Introduction -- 16.2 Multiple expressions for causal effect -- 16.3 Asymptotic variance of causal estimators -- 16.4 Comparison of causal estimators -- 16.4.1 Supplement C with L or not -- 16.4.2 Supplement L with C or not -- 16.4.3 Replace C with L or not
16.5 Discussion -- Acknowledgements -- Appendices -- 16.A Estimator given all X's recorded -- 16.B Derivations of asymptotic variances -- 16.B.1 Estimator given R, C and T recorded -- 16.B.2 Estimator given R, L and T recorded -- 16.C Expressions with correlation coefficients -- 16.C.1 Estimator given R, C and T recorded -- 16.C.2 Estimator given R, L and T recorded -- 16.C.3 Estimator given R, L, C and T recorded -- 16.D Derivation of I's -- 16.E Relation between ρ2rl and ρ2rl -- References -- 17 Time-varying confounding: Some practical considerations in a likelihood framework -- 17.1 Introduction -- 17.2 General setting -- 17.2.1 Notation -- 17.2.2 Observed data structure -- 17.2.3 Intervention strategies -- 17.2.4 Potential outcomes -- 17.2.5 Time-to-event outcomes -- 17.2.6 Causal estimands -- 17.3 Identifying assumptions -- 17.4 G-computation formula -- 17.4.1 The formula -- 17.4.2 Plug-in regression estimation -- 17.5 Implementation by Monte Carlo simulation -- 17.5.1 Simulating an end-of-study outcome -- 17.5.2 Simulating a time-to-event outcome -- 17.5.3 Inference -- 17.5.4 Losses to follow-up -- 17.5.5 Software -- 17.6 Analyses of simulated data -- 17.6.1 The data -- 17.6.2 Regimes to be compared -- 17.6.3 Parametric modelling choices -- 17.6.4 Results -- 17.7 Further considerations -- 17.7.1 Parametric model misspecification -- 17.7.2 Competing events -- 17.7.3 Unbalanced measurement times -- 17.8 Summary -- References -- 18 'Natural experiments' as a means of testing causal inferences -- 18.1 Introduction -- 18.2 Noncausal interpretations of an association -- 18.3 Dealing with confounders -- 18.4 'Natural experiments' -- 18.4.1 Genetically sensitive designs -- 18.4.2 Children of twins (CoT) design -- 18.4.3 Strategies to identify the key environmental risk feature -- 18.4.4 Designs for dealing with selection bias
18.4.5 Instrumental variables to rule out reverse causation
A state of the art volume on statistical causality Causality: Statistical Perspectives and Applications presents a wide-ranging collection of seminal contributions by renowned experts in the field, providing a thorough treatment of all aspects of statistical causality. It covers the various formalisms in current use, methods for applying them to specific problems, and the special requirements of a range of examples from medicine, biology and economics to political science. This book: Provides a clear account and comparison of formal languages, concepts and models for statistical causality.  Addresses examples from medicine, biology, economics and political science to aid the reader's understanding.  Is authored by leading experts in their field.  Is written in an accessible style. Postgraduates, professional statisticians and researchers in academia and industry will benefit from this book
Description based on publisher supplied metadata and other sources
Electronic reproduction. Ann Arbor, Michigan : ProQuest Ebook Central, 2020. Available via World Wide Web. Access may be limited to ProQuest Ebook Central affiliated libraries
Link Print version: Berzuini, Carlo Causality : Statistical Perspectives and Applications New York : John Wiley & Sons, Incorporated,c2012 9780470665565
Subject Estimation theory.;Causation.;Causality (Physics)
Electronic books
Alt Author Dawid, Philip
Bernardinell, Luisa
Record:   Prev Next