Meer over de auteurs

Mark J. van der Laan, Sherri Rose

Targeted Learning in Data Science

Name: Targeted Learning in Data Science
Author: Mark J. van der Laan

Causal Inference for Complex Longitudinal Studies

Gebonden Engels 2018 9783319653037

Verwachte levertijd ongeveer 9 werkdagen

132,99

In winkelwagen

Samenvatting

This textbook for graduate students in statistics, data science, and public health deals with the practical challenges that come with big, complex, and dynamic data. It presents a scientific roadmap to translate real-world data science applications into formal statistical estimation problems by using the general template of targeted maximum likelihood estimators. These targeted machine learning algorithms estimate quantities of interest while still providing valid inference. Targeted learning methods within data science area critical component for solving scientific problems in the modern age. The techniques can answer complex questions including optimal rules for assigning treatment based on longitudinal data with time-dependent confounding, as well as other estimands in dependent data structures, such as networks. Included in Targeted Learning in Data Science are demonstrations with soft ware packages and real data sets that present a case that targeted learning is crucial for the next generation of statisticians and data scientists. Th is book is a sequel to the first textbook on machine learning for causal inference, Targeted Learning, published in 2011.

Mark van der Laan, PhD, is Jiann-Ping Hsu/Karl E. Peace Professor of Biostatistics and Statistics at UC Berkeley. His research interests include statistical methods in genomics, survival analysis, censored data, machine learning, semiparametric models, causal inference, and targeted learning. Dr. van der Laan received the 2004 Mortimer Spiegelman Award, the 2005 Van Dantzig Award, the 2005 COPSS Snedecor Award, the 2005 COPSS Presidential Award, and has graduated over 40 PhD students in biostatistics and statistics.

Sherri Rose, PhD, is Associate Professor of Health Care Policy (Biostatistics) at Harvard Medical School. Her work is centered on developing and integrating innovative statistical approaches to advance human health. Dr. Rose’s methodological research focuses on nonparametric machine learning for causal inference and prediction. She co-leads the Health Policy Data Science Lab and currently serves as an associate editor for the Journal of the American Statistical Association and Biostatistics.

Specificaties

ISBN13:9783319653037

Taal:Engels

Bindwijze:gebonden

Uitgever:Springer International Publishing

Serie:Springer Series in Statistics

Lezersrecensies

Wees de eerste die een lezersrecensie schrijft!

Schrijf een recensie

Uw cijfer

?

Log in om te stemmen

Inhoudsopgave

Part I:  Introductory Chapters 1.  The Statistical Estimation Problem in Complex Longitudinal Data <ul><li>Data Science and Statistical Estimation </li><li>Roadmap for Causal Effect Estimation </li><li>Role of Targeted Learning in Data Science </li><li>Observed Data </li><li>Caussal Model and Causal target Quantity </li><li>Statistical Model </li><li>Statistical Target Parameter </li><li>Statistical Estimation Problem </li></ul> 2.  Longitudinal Causal Models <ul><li>Structural Causal Models </li><li>Causal Graphs / DAGs </li><li>Nonparametric Structural Equation Models </li></ul> 3.  Super Learner for Longitudinal Problems <ul><li>Ensemble Learning </li><li>Sequential Regression </li></ul> 4.  Longitudinal Targeted Maximum Likelihood Estimation (LTMLE) <ul><li>Step-by-Step Demonstration of LTMLE </li></ul>scalable inference="" for="" big="" data 5.  Understanding LTMLE <ul><li>Statistical Properties </li><li>Theoretical Background </li></ul> 6.  Why LTMLE? <ul><li>Landscape of Other Estimators </li><li>Comparison of Statistical Properties </li></ul>   Part II:  Additional Core Topics 7.  One-Step TMLE <ul><li>General Framework </li><li>Theoretical Results </li></ul> 8.  One-Step TMLE for the Effect Among the Treated <ul><li>Demonstration for Effect Among the Treated </li><li>Simulation Studies </li></ul> 9.  Online Targeted Learning <ul><li>Batched Streaming Data </li><li>Online and One-Step Estimator </li><li>Theoretical Considerations </li></ul> 10.  Networks <ul><li>General Statistical Framework </li><li>Causal Model for Network Da</li></ul>ta <li>Counterfactual Mean Under Stochastic Intervention on the Network </li><li>Development of TMLE for Networks </li><li>Inference </li> 11. Application to Networks <ul><li>Differing Network Structures </li><li>Realistic Network Examples (e.g., effect of vaccination) </li><li>R Package Implementation of TMLE </li></ul> 12. Targeted Estimation of the Nuisance Parameter <ul><li>Asymptotic Linearity </li><li>IPW </li><li>TMLE </li></ul> 13. Sensitivity Analyses<ul><li>General Nonparametric Approach to Sensitivity Analysis </li><li>Measurement Error </li><li>Unmeasured Confounding </li><li>Informative Missingness of the Outcome </li><li>FDA Meta-Analysis </li></ul>   Part III:  Randomized Trials 14. Community Randomized Trials for Small Samples <ul><li>Introduction of SEARCH Community Rando</li></ul>mized Trial <li>Adaptive Pair Matching </li><li>Data-Adaptive Selection of Covariates for Small Samples </li><li>TMLE Using Super Learning for Small Samples </li><li>Inference </li> 15. Sample Average Treatment Effect in a CRT <ul><li>Introduction of the Parameter </li><li>Effect for the Observed Communities </li><li>Inference </li></ul> 16. Application to Clinical Trial Survival Data <ul><li>Introduction of the Survival Parameter </li><li>Censoring </li><li>Treatment-Specific Survival Function </li></ul> 17. Application to Pandora Music Data <ul><li>Effect of Pandora Streaming on Music Sales </li><li>Application of TMLE </li></ul> 18. Causal Effect Transported Across Sites <ul><li>Intent-to-Treat ATE </li><li>Complier ATE </li><li>Incomplete Data </li><li>Moving to Opportunity Trial </li></ul>   Part IV:  Observational Longitudinal Data 19. Super Learning in the ICU <ul><li>ICU Prediction Problem </li><li>Super Learning Algorithm</li></ul> <ul><li>Defining Stochastic Interventions </li><li>Dependence on True Treatment Mechanisms </li><li>Continuous Exposure </li><li>Air Pollution Data Example </li></ul> 21. Stochastic Multiple-Time-Point Interventions on Monitoring and Treatment <ul><li>Defining Stochastic Interventions for Multiple-Time Points </li><li>Introduction of Monitoring Problem </li><li>Non-direct Effect Assumption of Monitoring </li><li>Dynamic Treatment </li><li>Diabetes Data Example </li></ul> 22. Collaborative LTMLE <ul><li>Collaborative LTMLE Framework </li><li>Breastfeeding Data Example </li></ul>   Part V:  Optimal Dynamic Regimes 23. Targeted Adaptive Designs Learning the Optimal Dynamic Treatment <ul><li>Group-Sequential Adaptive Designs </li><li>Multiple Bandit Problem </li><li>Treatment Allocation Learning from Past Data </li><li>Mean Outcome Under the Optimal Treatment </li><li>Martingale Theory </li><li>Inference </li></ul> 24. Targeted Learning of the Optimal Dynamic Treatment <ul><li>Super Learning for Discovering the Optimal Dynamic rule </li><li>Different Loss Functions </li><li>TMLE for the Counterfactual Mean </li><li>Statistical Inference for  the Mean Outcome Under the Optimal Rule </li></ul> 25. Optimal Dynamic Treatments Under Resource Constraints<ul><li>Constrained Optimal Dynamic Treatment </li><li>Super Learning of the Constrained Optimal Dynamic Regime </li><li>TMLE of the Counterfactual Mean Under the Constrained </li></ul>Optimal Dynamic Regime   Part VI:  Computing 26. ltmle() for R <ul><li>Introduction to the ltmle() R Package </li><li>Demonstration of the ltmle() R Package </li></ul> 27. Scaled Super Learner for R Introduction to the H2O Environment<ul><li>R Package </li><li>Subsemble </li></ul> 28. Scaling CTMLE for Julia <ul><li>Scaling Computing of CTMLE in Julia </li><li>Pharmacoepidemiology Example </li></ul>   Part VII:  Special Topics 29. Data-Adaptive Target Parameters <ul><li>Definition of Parameter </li><li>Examples of Data-Adaptive Target Parameters as Arise in Data Mining </li><li>Estimators of the Data-Adaptive Target Parameters Using Sample Splitting </li><li>Estimators of the Data-Adaptive Target Parameters Without Sample Splitting </li><li>Cross-Validated TMLE of the Data-Adap</li></ul>tive Target Parameters 30. Double Robust Inference for LTMLE <ul><li>The Challenge of Double Robust Inference for Double Robust Estimators </li> 31. Higher-Order TMLE<ul><li>Higher-Order Pathwise Differentiable Target Parameters </li><li>Higher-Order TMLE </li><li>Kth Order Remainder </li><li>Parameters Not Second-Order Pathwise Differentiable </li><li>Second-Order U Statistics </li><li>Approximate Second-Order Influence Function </li><li>Approximate Second-Order TMLE </li></ul>   Appendices A.  Online Targeted Learning Theory B.  Computerization of the Calculation of Efficient Influence Curve</ul> C.  TMLE Applied to Capture/Recapture D.  TMLE for High Dimensional Linear Regression E.  TMLE of Causal Effect Based on Observing a Single Time Series

Managementboek Top 100

Bekijk de volledige Managementboek Top 100

Uw winkelwagen

Targeted Learning in Data Science

Causal Inference for Complex Longitudinal Studies

Samenvatting

Specificaties

Lezersrecensies

Inhoudsopgave

Managementboek Top 100

Rubrieken

Populaire producten

Personen

Trefwoorden