Publisher Description

The statistics profession is at a unique point in history. The need for valid statistical tools is greater than ever; data sets are massive, often measuring hundreds of thousands of measurements for a single subject. The field is ready to move towards clear objective benchmarks under which tools can be evaluated. Targeted learning allows (1) the full generalization and utilization of cross-validation as an estimator selection tool so that the subjective choices made by humans are now made by the machine, and (2) targeting the fitting of the probability distribution of the data toward the target parameter representing the scientific question of interest.
 
This book is aimed at both statisticians and applied researchers interested in causal inference and general effect estimation for observational and experimental data. Part I is an accessible introduction to super learning and the targeted maximum likelihood estimator, including related concepts necessary to understand and apply these methods. Parts II-IX handle complex data structures and topics applied researchers will immediately recognize from their own research, including time-to-event outcomes, direct and indirect effects, positivity violations, case-control studies, censored data, longitudinal data, and genomic studies.

Back Cover

The statistics profession is at a unique point in history. The need for valid statistical tools is greater than ever; data sets are massive, often measuring hundreds of thousands of measurements for a single subject. The field is ready to move towards clear objective benchmarks under which tools can be evaluated. Targeted learning allows (1) the full generalization and utilization of cross-validation as an estimator selection tool so that the subjective choices made by humans are now made by the machine, and (2) targeting the fitting of the probability distribution of the data toward the target parameter representing the scientific question of interest. This book is aimed at both statisticians and applied researchers interested in causal inference and general effect estimation for observational and experimental data. Part I is an accessible introduction to super learning and the targeted maximum likelihood estimator, including related concepts necessary to understand and apply these methods. Parts II-IX handle complex data structures and topics applied researchers will immediately recognize from their own research, including time-to-event outcomes, direct and indirect effects, positivity violations, case-control studies, censored data, longitudinal data, and genomic studies. "Targeted Learning, by Mark J. van der Laan and Sherri Rose, fills a much needed gap in statistical and causal inference. It protects us from wasting computational, analytical, and data resources on irrelevant aspects of a problem and teaches us how to focus on what is relevant - answering questions that researchers truly care about." -Judea Pearl, Computer Science Department, University of California, Los Angeles "In summary, this book should be on the shelf of every investigator who conducts observational research and randomized controlled trials. The concepts and methodology are foundational for causal inference and at the same time stay true to what the data at hand can say about the questions that motivate their collection." -Ira B. Tager, Division of Epidemiology, University of California, Berkeley

Table of Contents

Models, Inference, and Truth.- The Open Problem.- Defining the Model and Parameter.- Super Learning.- Introduction to TMLE.- Understanding TMLE.- Why TMLE?.- Bounded Continuous Outcomes.- Direct Effects and Effect Among the Treated.- Marginal Structural Models.- Positivity.- Robust Analysis of RCTs Using Generalized Linear Models.- Targeted ANCOVA Estimator in RCTs.- Independent Case-Control Studies.- Why Match? Matched Case-Control Studies.- Nested Case-Control Risk Score Prediction.- Super Learning for Right-Censored Data.- RCTs with Time-to-Event Outcomes.- RCTs with Time-to-Event Outcomes and Effect Modification Parameters.- C-TMLE of an Additive Point Treatment Effect.- C-TMLE for Time-to-Event Outcomes.- Propensity-Score-Based Estimators and C-TMLE.- Targeted Methods for Biomarker Discovery.- Finding Quantitative Trait Loci Genes.- Case Study: Longitudinal HIV Cohort Data.- Probability of Success of an In Vitro Fertilization Program.- Individualized Antiretroviral Initiation Rules.- Cross-Validated Targeted Minimum-Loss-Based Estimation.- Targeted Bayesian Learning.- TMLE in Adaptive Group Sequential Covariate Adjusted RCTs.- Foundations of TMLE.- Introduction to R Code Implementation.

Review

From the reviews:"This book is a timely fit and is expected to draw much attention from researchers in the field of causal inference. The book explains the concept of targeted learning, which is an enhanced procedure for estimating targeted causal estimands under the potential outcome framework. … Excellent summaries of complex estimation procedures and methods are ubiquitous, which will be helpful for the nontechnical readers of the book. … This book appears to be a useful reference for Ph.D. students in biostatistics programs." (Joseph Kang, Journal of the American Statistical Association, June, 2013)

Long Description

The statistics profession is at a unique point in history. The need for valid statistical tools is greater than ever; data sets are massive, often measuring hundreds of thousands of measurements for a single subject. The field is ready to move towards clear objective benchmarks under which tools can be evaluated. Targeted learning allows (1) the full generalization and utilization of cross-validation as an estimator selection tool so that the subjective choices made by humans are now made by the machine, and (2) targeting the fitting of the probability distribution of the data toward the target parameter representing the scientific question of interest. This book is aimed at both statisticians and applied researchers interested in causal inference and general effect estimation for observational and experimental data. Part I is an accessible introduction to super learning and the targeted maximum likelihood estimator, including related concepts necessary to understand and apply these methods. Parts II-IX handle complex data structures and topics applied researchers will immediately recognize from their own research, including time-to-event outcomes, direct and indirect effects, positivity violations, case-control studies, censored data, longitudinal data, and genomic studies.

Review Quote

From the reviews: "This book is a timely fit and is expected to draw much attention from researchers in the field of causal inference. The book explains the concept of targeted learning, which is an enhanced procedure for estimating targeted causal estimands under the potential outcome framework. ... Excellent summaries of complex estimation procedures and methods are ubiquitous, which will be helpful for the nontechnical readers of the book. ... This book appears to be a useful reference for Ph.D. students in biostatistics programs." (Joseph Kang, Journal of the American Statistical Association, June, 2013)

Feature

Establishes causal inference methodology that incorporates the benefits of machine learning with statistical inference Presentation combines accessibility with the method's rigorous grounding in statistical theory Demonstrates targeted learning in epidemiological, medical, and genomic experimental and observational studies that include informative dropout, missingness, time-dependent confounding, and case-control sampling

Details