Atjaunināt sīkdatņu piekrišanu

Statistical Methods for Dynamic Treatment Regimes: Reinforcement Learning, Causal Inference, and Personalized Medicine 2013 ed. [Hardback]

3.67/5 (3 ratings by Goodreads)
,
  • Formāts: Hardback, 204 pages, height x width: 235x155 mm, weight: 4557 g, XVI, 204 p., 1 Hardback
  • Sērija : Statistics for Biology and Health 76
  • Izdošanas datums: 23-Jul-2013
  • Izdevniecība: Springer-Verlag New York Inc.
  • ISBN-10: 1461474272
  • ISBN-13: 9781461474272
Citas grāmatas par šo tēmu:
  • Hardback
  • Cena: 100,46 €*
  • * ši ir gala cena, t.i., netiek piemērotas nekādas papildus atlaides
  • Standarta cena: 118,19 €
  • Ietaupiet 15%
  • Grāmatu piegādes laiks ir 3-4 nedēļas, ja grāmata ir uz vietas izdevniecības noliktavā. Ja izdevējam nepieciešams publicēt jaunu tirāžu, grāmatas piegāde var aizkavēties.
  • Daudzums:
  • Ielikt grozā
  • Piegādes laiks - 4-6 nedēļas
  • Pievienot vēlmju sarakstam
Statistical Methods for Dynamic Treatment Regimes: Reinforcement Learning, Causal Inference, and Personalized Medicine 2013 ed.Palielināt
  • Formāts: Hardback, 204 pages, height x width: 235x155 mm, weight: 4557 g, XVI, 204 p., 1 Hardback
  • Sērija : Statistics for Biology and Health 76
  • Izdošanas datums: 23-Jul-2013
  • Izdevniecība: Springer-Verlag New York Inc.
  • ISBN-10: 1461474272
  • ISBN-13: 9781461474272
Citas grāmatas par šo tēmu:
Permanent link: https://www.kriso.lv/db/9781461474272.html
Keywords:

Statistical Methods for Dynamic Treatment Regimes shares state of the art of statistical methods developed to address questions of estimation and inference for dynamic treatment regimes, a branch of personalized medicine. This volume demonstrates these methods with their conceptual underpinnings and illustration through analysis of real and simulated data. These methods are immediately applicable to the practice of personalized medicine, which is a medical paradigm that emphasizes the systematic use of individual patient information to optimize patient health care. This is the first single source to provide an overview of methodology and results gathered from journals, proceedings, and technical reports with the goal of orienting researchers to the field. The first chapter establishes context for the statistical reader in the landscape of personalized medicine. Readers need only have familiarity with elementary calculus, linear algebra, and basic large-sample theory to use this text. Throughout the text, authors direct readers to available code or packages in different statistical languages to facilitate implementation. In cases where code does not already exist, the authors provide analytic approaches in sufficient detail that any researcher with knowledge of statistical programming could implement the methods from scratch. This will be an important volume for a wide range of researchers, including statisticians, epidemiologists, medical researchers, and machine learning researchers interested in medical applications. Advanced graduate students in statistics and biostatistics will also find material in Statistical Methods for Dynamic Treatment Regimes to be a critical part of their studies.

Recenzijas

From the reviews:

"Overall, the book provides an excellent reviewof DTRs up to date. After finishing reading the book, I planned to create a post-graduate seminar course on this topic using this book as a textbook. I enthusiastically recommend this book. This book will be a valuable reference for anyone interested and involved in research on personalized medicine." (Hyonggin An, Journal of Agricultural, Biological, and Environmental Statistics, April, 2015)

The intended audience includes physicians, clinical researchers, physicians in training, statisticians, and medical students, as well as masters and doctoral students in the field of biostatistics and epidemiology and computer scientists. This book provides a concise summary of the key findings in the statistical literature of dynamic treatment regimes. The simple language and well-organized chapters are unsurpassed attributes of this book. It will be an exceptional resource for quick review. (Parthiv Amin, Doodys Book Reviews, November, 2013)

1 Introduction
1(8)
1.1 Evidence-Based Personalized Medicine for Chronic Diseases
1(1)
1.2 Personalized Medicine and Medical Decision Making
2(5)
1.2.2 Single-stage Decision Problems in Personalized Medicine
3(1)
1.2.2 Multi-stage Decisions and Dynamic Treatment Regimes
4(3)
1.3 Outline of the Book
7(2)
2 The Data: Observational Studies and Sequentially Randomized Trials
9(22)
2.1 Longitudinal Observational Studies
9(6)
2.1.1 The Potential Outcomes Framework
10(1)
2.1.1 Time-Varying Confounding and Mediation
11(2)
2.1.1 Necessary Assumptions
13(2)
2.2 Examples of Longitudinal Observational Studies
15(3)
2.2.2 Investigating Warfarin Dosing Using Hospital Data
16(1)
2.2.2 Investigating Epoetin Therapy Using the United States Renal Data System
16(1)
2.2.2 Estimating Optimal Breastfeeding Strategies Using Data from a Randomized Encouragement Trial
17(1)
2.3 Sequentially Randomized Studies
18(10)
2.3.3 SMART Versus a Series of Single-stage Randomized Trials
20(1)
2.3.3 Design Properties
21(4)
2.3.3 Practical Considerations
25(1)
2.3.3 SMART Versus Other Designs
26(2)
2.4 Examples of Sequentially Randomized Studies
28(2)
2.4.4 Project Quit - Forever Free: A Smoking Cessation Study
28(1)
2.4.4 STAR*D: A Study of Depression
29(1)
2.5 Discussion
30(1)
3 Statistical Reinforcement Learning
31(22)
3.1 Multi-stage Decision Problems
31(1)
3.2 Reinforcement Learning: A Conceptual Overview
32(3)
3.3 A Probabilistic Framework
35(3)
3.4 Estimation of Optimal DTRs by Modeling Conditional Means
38(7)
3.4.4 Q-learning with Linear Models
39(2)
3.4.4 Why Move Through Stages?
41(2)
3.4.4 Analysis of Smoking Cessation Data: An Illustration
43(2)
3.5 Q-learning Using Observational Data
45(6)
3.6 Discussion
51(2)
4 Semi-parametric Estimation of Optimal DTRs by Modeling Contrasts of Conditional Mean Outcomes
53(26)
4.1 Structural Nested Mean Models
53(4)
4.1.1 Special Cases of Optimal SNMMs
55(1)
4.1.1 Standard SNMMs
56(1)
4.2 Model Parameterizations and Optimal Rules
57(3)
4.3 G-estimation
60(6)
4.3.3 More Efficient G-estimation
62(1)
4.3.3 Recursive G-estimation
63(1)
4.3.3 G-estimation Versus OLS Regression for a One-Stage Problem
64(1)
4.3.3 Q-learning and G-estimation
65(1)
4.4 Regret-Based Methods of Estimation
66(11)
4.4.4 Iterative Minimization of Regrets
67(4)
4.4.4 A-Learning
71(1)
4.4.4 Regret-Regression
71(2)
4.4.4 Occlusion Therapy for Amblyopia: An Illustration
73(3)
4.4.4 Simulation of a Two-Stage Occlusion Trial for Treatment of Amblyopia
76(1)
4.5 Discussion
77(2)
5 Estimation of Optimal DTRs by Directly Modeling Regimes
79(22)
5.1 Estimating the Value of an Arbitrary Regime: Inverse Probability Weighting
80(3)
5.2 Marginal Structural Models and Weighting Methods
83(11)
5.2.2 MSMs for Static Treatment Regimes
83(1)
5.2.2 MSMs for Dynamic Treatment Regimes
84(4)
5.2.2 Simulation of a DTR MSM Analysis to Determine the Optimal Treatment Threshold
88(2)
5.2.2 Treatment for Schizophrenia: An Illustration
90(4)
5.3 A Classification Approach to Estimating DTRs
94(3)
5.3.3 Contrast-Weighted Classification
95(1)
5.3.3 Outcome Weighted Learning
96(1)
5.4 Assessing the Merit of an Estimated Regime
97(2)
5.5 Discussion
99(2)
6 G-computation: Parametric Estimation of Optimal DTRs
101(12)
6.1 Frequentist G-computation
101(6)
6.1.1 Applications and Implementation of G-computation
103(1)
6.1.1 Breastfeeding and Vocabulary: An Illustration
104(3)
6.2 Bayesian Estimation of DTRs
107(5)
6.2.2 Approach and Applications
107(2)
6.2.2 Assumptions as Viewed in a Bayesian Framework
109(1)
6.2.2 Breastfeeding and Vocabulary: An Illustration, Continued
109(3)
6.3 Discussion
112(1)
7 Estimation of DTRs for Alternative Outcome Types
113(14)
7.1 Trading Off Multiple Rewards: Multi-dimensional and Compound Outcomes
113(1)
7.2 Estimating DTRs for Time-to-Event Outcomes with Q-learning
114(6)
7.2.2 Simple Q-learning for Survival Data: IPW in Sequential AFT Models
114(1)
7.2.2 Q-learning with Support Vector Regression for Censored Survival Data
115(5)
7.3 Q-learning of DTRs for Discrete Outcomes
120(2)
7.4 Inverse Probability Weighted Estimation for Censored or Discrete Outcomes and Stochastic Treatment Regimes
122(2)
7.5 Estimating a DTR for a Binary Outcome Using a Likelihood Approach
124(1)
7.6 Discussion
125(2)
8 Inference and Non-regularity
127(42)
8.1 Inference for the Parameters Indexing the Optimal Regime Under Regularity
128(8)
8.1.1 A Brief Review of Variances for Estimating Equations
129(2)
8.1.1 Asymptotic Variance for Q-learning Estimators
131(2)
8.1.1 Asymptotic Variance for G-estimators
133(2)
8.1.1 Projection Confidence Intervals
135(1)
8.2 Exceptional Laws and Non-regularity of the Parameters Indexing the Optimal Regime
136(3)
8.2.2 Non-regularity in Q-learning
138(1)
8.2.2 Non-regularity in G-estimation
139(1)
8.3 Threshold Estimators with the Usual Bootstrap
139(6)
8.3.3 The Hard-Threshold Estimator
140(1)
8.3.3 The Soft-Threshold Estimator
141(2)
8.3.3 Analysis of Smoking Cessation Data: An Illustration, Continued
143(2)
8.4 Penalized Q-learning
145(3)
8.5 Double Bootstrap Confidence Intervals
148(1)
8.6 Adaptive Bootstrap Confidence Intervals
149(2)
8.7 m-out-of-n Bootstrap Confidence Intervals
151(3)
8.8 Simulation Study
154(6)
8.9 Analysis of STAR*D Data: An Illustration
160(4)
8.9.9 Background and Study Details
160(1)
8.9.9 Analysis
161(1)
8.9.9 Results
162(2)
8.10 Inference About the Value of an Estimated DTR
164(2)
8.11 Bayesian Estimation in Non-regular Settings
166(1)
8.12 Discussion
166(3)
9 Additional Considerations and Final Thoughts
169(12)
9.1 Variable Selection
169(5)
9.1.1 Penalized Regression
170(1)
9.1.1 Variable Ranking by Qualitative Interactions
171(2)
9.1.1 Stepwise Selection
173(1)
9.2 Model Checking via Residual Diagnostics
174(3)
9.3 Discussion and Concluding Remarks
177(4)
Glossary 181(4)
References 185(18)
Index 203
Bibhas Chakraborty is an Assistant Professor of Biostatistics at the Mailman School of Public Health, Columbia University. His primary research interests lie in dynamic treatment regimes, machine learning and data mining including reinforcement learning, causal inference, and design and analysis of clinical trials. He received a Bachelors degree from the University of Calcutta, a Masters degree from the Indian Statistical Institute, and a Ph.D. in Statistics from the University of Michigan. He is the recipient of the Calderone Research Prize for Junior Faculty from the Mailman School of Public Health, Columbia University, in 2011.

Erica Moodie is an Associate Professor of Biostatistics in the Department of Epidemiology, Biostatistics, and Occupational Health at McGill University. Her main interests lie in causal inference and longitudinal data with a focus on methods for HIV research. She is an Associate Editor of The International Journal of Biostatistics and Journal of Causal Inference. She received a bachelor's degree in Mathematics and Statistics from the University of Winnipeg, an M.Phil. in Epidemiology from the University of Cambridge, and a Ph.D. in Biostatistics from the University of Washington. She is the recipient of a Natural Sciences and Engineering Research Council University Faculty Award.