Muutke küpsiste eelistusi

E-raamat: Statistical Methods for Mediation, Confounding and Moderation Analysis Using R and SAS [Taylor & Francis e-raamat]

(Louisiana State University, LA, USA),
  • Formaat: 294 pages, 18 Tables, black and white; 100 Line drawings, black and white; 100 Illustrations, black and white
  • Sari: Chapman & Hall/CRC Biostatistics Series
  • Ilmumisaeg: 21-Mar-2022
  • Kirjastus: Chapman & Hall/CRC
  • ISBN-13: 9780429346941
Teised raamatud teemal:
  • Taylor & Francis e-raamat
  • Hind: 216,96 €*
  • * hind, mis tagab piiramatu üheaegsete kasutajate arvuga ligipääsu piiramatuks ajaks
  • Tavahind: 309,94 €
  • Säästad 30%
Statistical Methods for Mediation, Confounding and Moderation Analysis Using R and SASSuurem pilt
  • Formaat: 294 pages, 18 Tables, black and white; 100 Line drawings, black and white; 100 Illustrations, black and white
  • Sari: Chapman & Hall/CRC Biostatistics Series
  • Ilmumisaeg: 21-Mar-2022
  • Kirjastus: Chapman & Hall/CRC
  • ISBN-13: 9780429346941
Teised raamatud teemal:
Taylor & Francis e-raamatudRohkem infot Taylor & Francis e-raamatute kohta
Raamatu kodulehekülg: https://www.taylorfrancis.com/books/9780429346941
Third-variable effect refers to the effect transmitted by third-variables that intervene in the relationship between an exposure and a response variable. Differentiating between the indirect effect of individual factors from multiple third-variables is a constant problem for modern researchers.

Statistical Methods for Mediation, Confounding and Moderation Analysis Using R and SAS introduces general definitions of third-variable effects that are adaptable to all different types of response (categorical or continuous), exposure, or third-variables. Using this method, multiple third- variables of different types can be considered simultaneously, and the indirect effect carried by individual third-variables can be separated from the total effect. Readers of all disciplines familiar with introductory statistics will find this a valuable resource for analysis.

Key Features:





Parametric and nonparametric method in third variable analysis Multivariate and Multiple third-variable effect analysis Multilevel mediation/confounding analysis Third-variable effect analysis with high-dimensional data Moderation/Interaction effect analysis within the third-variable analysis R packages and SAS macros to implement methods proposed in the book
Preface xiii
Symbols xv
1 Introduction
1(8)
1.1 Types of Third-Variable Effects
1(2)
1.2 Motivate Examples for Making Inferences on Third-Variable Effects
3(3)
1.2.1 Evaluate Policies and Interventions
3(1)
1.2.2 Explore Health Disparities
4(1)
1.2.3 Exam the Trend of Disparities
5(1)
1.3 Organization of the Book
6(3)
2 A Review of Third-Variable Effect Inferences
9(10)
2.1 The General Linear Model Framework
9(6)
2.1.1 Baron and Kenny Method to Identify a Third-Variable Effect
9(2)
2.1.2 The Coefficient-Difference Method
11(1)
2.1.3 The Coefficient-Product Method
12(1)
2.1.4 Categorical Third-Variables
13(1)
2.1.5 Generalized Linear Model for the Outcome
14(1)
2.1.6 Cox Proportional Hazard Model for Time-to-Event Outcome
15(1)
2.2 The Counterfactual Framework
15(4)
2.2.1 Binary Exposures
16(1)
2.2.2 Continuous Exposure
17(1)
2.2.3 Discussion
17(2)
3 Advanced Statistical Modeling and Machine Learning Methods Used in the Book
19(24)
3.1 Bootstrapping
19(3)
3.1.1 An Illustration of Bootstrapping
19(1)
3.1.2 Bootstrapping for Linear Regression
20(2)
3.2 Elastic Net
22(5)
3.2.1 Ridge Regression and LASSO
24(2)
3.2.2 Elastic Net
26(1)
3.3 Multiple Additive Regression Trees
27(8)
3.3.1 Classification and Regression Tree
28(2)
3.3.2 MART Algorithm
30(1)
3.3.3 Improvement of MART
31(1)
3.3.4 A Simulation Example
31(2)
3.3.5 Interpretation Tools for MART
33(2)
3.4 Generalized Additive Model
35(8)
3.4.1 Generalized Additive Model
35(2)
3.4.2 Smoothing Spline
37(1)
3.4.3 Revisit the Simulation Example
37(6)
4 The General Third-Variable Effect Analysis Method
43(18)
4.1 Notations
44(1)
4.2 Definitions of Third-Variable Effects
44(3)
4.2.1 Total Effect
45(1)
4.2.2 Direct and Indirect Effects
45(2)
4.2.3 Relative Effect
47(1)
4.3 Third-Variable Effect Analysis with Generalized Linear Models
47(9)
4.3.1 Multiple Third-Variable Analysis in Linear Regressions
47(2)
4.3.2 Multiple Third-Variable Analysis in Logistic Regression
49(1)
4.3.2.1 When M is Binary
50(2)
4.3.2.2 When M is Multi-categorical
52(1)
4.3.2.3 Delta Method to Estimate the Variances
53(3)
4.4 Algorithms of Third-Variable Effect Analysis with General Predictive Models for Binary Exposure
56(2)
4.5 Algorithms of Third-Variable Effect Analysis with General Predictive Models for Continuous Exposure
58(3)
5 The Implementation of General Third-Variable Effect Analysis Method
61(32)
5.1 The R Package mma
61(10)
5.1.1 Identification of Potential Mediators/Confounders and Organization of Data
62(4)
5.1.2 Third-Variable Effect Estimates
66(2)
5.1.3 Statistical Inference on Third-Variable Effect Analysis
68(3)
5.2 SAS Macros
71(6)
5.2.1 Running R in SAS
72(2)
5.2.2 Macros to Call the data.org Function
74(1)
5.2.3 Macros to Call the med Function
75(1)
5.2.4 Macros to Call the boot.med Function
76(1)
5.2.5 Macros to Call the plot Function
77(1)
5.3 Examples and Simulations on General Third-Variable Effect Analysis
77(16)
5.3.1 Pattern of Care Study
77(1)
5.3.2 To Explore the Racial Disparity in Breast Cancer Mortality Rate
78(5)
5.3.3 To Explore the Racial Disparity in Breast Cancer Survival Rate
83(4)
5.3.4 Simulation Study
87(2)
5.3.4.1 Empirical Bias
89(1)
5.3.4.2 Type I Error Rate and Power
89(4)
6 Assumptions for the General Third-Variable Analysis
93(30)
6.1 Assumption 1: No-Unmeasured-Confounder for the Exposure-Outcome Relationship
94(7)
6.1.1 On the Direct Effect
95(2)
6.1.2 On the Indirect Effect of M
97(1)
6.1.3 On the Total Effect
97(2)
6.1.4 Summary and the Correct Model
99(2)
6.2 Assumption 2: No-Unmeasured-Confounder for the Exposure-Third Variable Relationship
101(4)
6.2.1 On the Direct Effect
102(1)
6.2.2 On the Indirect Effect of M
103(1)
6.2.3 On the Total Effect
104(1)
6.2.4 Summary and the Correct Model
105(1)
6.3 Assumption 3: No-Unmeasured-Confounder for the Third Variable-Outcome Relationship
105(9)
6.3.1 On the Total Effect
108(1)
6.3.2 On the Direct Effect
109(3)
6.3.3 On the Indirect Effect of M
112(2)
6.3.4 Summary and the Correct Model
114(1)
6.4 Assumption 4: Any Third-Variable Mi is not Causally Prior to Other Third-Variables in M_i
114(9)
6.4.1 On the Direct Effect
115(2)
6.4.2 On the Indirect Effect of M
117(2)
6.4.3 On the Total Effect
119(3)
6.4.4 Conclusion
122(1)
7 Multiple Exposures and Multivariate Responses
123(20)
7.1 Multivariate Multiple TVEA
123(3)
7.1.1 Non/Semi-Parametric TVEA for Multi-Categorical Exposures
124(2)
7.1.2 Non/Semi-Parametric TVEA for Multiple Continuous Exposures
126(1)
7.2 Confidence Ball for Estimated Mediation Effects
126(2)
7.2.1 A Simulation Study to Check the Coverage Probability of the Confidence Ball
127(1)
7.3 The R Package mma
128(4)
7.4 Racial and Ethnic Disparities in Obesity and BMI
132(11)
7.4.1 Variables
133(2)
7.4.2 Disparities
135(1)
7.4.3 Descriptive Analysis
136(2)
7.4.4 Results on Racial Disparities
138(2)
7.4.5 Results on Ethnic Disparities
140(3)
8 Regularized Third-Variable Effect Analysis for High-Dimensional Dataset
143(36)
8.1 Regularized Third-Variable Analysis in Linear Regression Setting
144(1)
8.2 Computation: The Algorithm to Estimate Third-variable Effects with Generalized Linear Models
145(3)
8.3 The R Package: mmabig
148(16)
8.3.1 Simulate a Dataset
148(1)
8.3.2 Function data.org.big
149(1)
8.3.2.1 Univariate Exposure and Univariate Outcome
149(3)
8.3.2.2 Survival Outcome
152(1)
8.3.2.3 Multivariate Predictors and/or Outcomes
153(3)
8.3.3 Function med.big
156(2)
8.3.4 Function mma.big
158(4)
8.3.5 Generic Functions
162(2)
8.3.6 Call mmabig from SAS
164(1)
8.4 Sensitivity and Specificity Analysis
164(3)
8.5 Simulations to Illustrate the Use of the Method
167(9)
8.5.1 X - TV Relationship is Nonlinear
169(2)
8.5.2 TV - Y Relationship is Nonlinear
171(2)
8.5.3 When Third-Variables are Highly Correlated
173(3)
8.6 Explore Racial Disparity in Breast Cancer Survival
176(3)
9 Interaction/Moderation Analysis with Third-Variable Effects
179(24)
9.1 Inference on Moderation Effect with Third-Variable Effect Analysis
180(4)
9.1.1 Types of Interaction/Moderation Effects
180(1)
9.1.2 Moderation Effect Analysis with MART
181(3)
9.1.3 The R Package mma
184(1)
9.2 Illustration of Moderation Effects
184(11)
9.2.1 Direct Moderation
184(4)
9.2.2 Exposure-Moderated TVE
188(3)
9.2.3 Third-Variable-Moderated TVE
191(4)
9.3 Explore the Trend of Racial Disparity in ODX Utilization among Breast Cancer Patients
195(7)
9.3.1 Data Description
196(2)
9.3.2 Third-Variable Effects and the Trend
198(4)
9.4 Conclusions
202(1)
10 Third-Variable Effect Analysis with Multilevel Additive Models
203(22)
10.1 Third-Variable Analysis with Multilevel Additive Models
204(7)
10.1.1 Definitions of Third-Variable Effects with Data of Two Levels
204(3)
10.1.2 Multilevel Additive Models
207(1)
10.1.3 Third-Variable Effects with Multilevel Additive Model
208(2)
10.1.4 Bootstrap Method for Third-Variable Effect Inferences
210(1)
10.2 The mlma R Package
211(11)
10.2.1 A Simulated Dataset
211(2)
10.2.2 Data Transformation and Organization
213(1)
10.2.3 Multilevel Third-Variable Analysis
214(1)
10.2.3.1 The mlma Function
214(1)
10.2.3.2 The Summary Function for Multilevel Third-Variable Analysis
215(2)
10.2.3.3 The Plot Function for the mlma Object
217(3)
10.2.4 Make Inferences on Multilevel Third-Variable Effect
220(2)
10.3 Explore the Racial Disparity in Obesity
222(3)
11 Bayesian Third-Variable Effect Analysis
225(20)
11.1 Why Bayesian Method?
225(2)
11.2 Continuous Exposure Variable
227(14)
11.2.1 Continuous Outcome and Third-Variables
227(1)
11.2.1.1 Method 1: Functions of Estimated Coefficients
228(2)
11.2.1.2 Method 2: Product of Partial Differences
230(2)
11.2.1.3 Method 3: A Resampling Method
232(2)
11.2.2 Different Format of Outcome and Third-Variables
234(1)
11.2.2.1 Outcomes of Different Format
234(1)
11.2.2.2 Binary Third-Variables
235(2)
11.2.2.3 Categorical Third-Variables
237(4)
11.3 Binary Exposure Variable
241(3)
11.4 Multiple Exposure Variables and Multivariate Outcomes
244(1)
12 Other Issues
245(10)
12.1 Explaining Third-Variable Effects
245(2)
12.2 Power Analysis and Sample Sizes
247(3)
12.2.1 Linear Models
247(2)
12.2.2 Simulation Method
249(1)
12.3 Sequential Third-Variable Analysis and Third-Variable Analysis with Longitudinal Data
250(5)
Appendices 255(12)
Bibliography 267(8)
Index 275
Qingzhao Yu is Professor in Biostatistics, Louisiana State University Health Sciences Center.

Bin Li is Associate Professor in Statistics, Louisiana State University.
Püsilink: https://www.kriso.ee/db/9780429346941_pe.html
Märksõnad: