Statistical Advances in the Biomedical Sciences -- Contents -- Preface -- Acknowledgments -- Contributors -- PART I CLINICAL TRIALS -- 1. Phase I Clinical Trials -- 1.1 Introduction -- 1.2 Phase I Trials in Healthy Volunteers -- 1.3 Phase I Trials with Toxic Outcomes Enrolling Patients -- 1.3.1 Parametric versus Nonparametric designs -- 1.3.2 Markovian-Motivated Up-and-Down Designs -- 1.3.3 Isotonic Designs -- 1.3.4 Bayesian Designs -- 1.3.5 Time-to-Event Design Modifications -- 1.4 Other Design Problems in Dose Finding -- 1.5 Concluding Remarks -- 2. Phase II Clinical Trials -- 2.1 Introduction -- 2.1.1 Background -- 2.1.2 The Role of Phase II Clinical Trials in Clinical Evaluation of a Novel Therapeutic Agent -- 2.1.3 Phase II Clinical Trial Designs -- 2.2 Frequentist Methods in Phase II Clinical Trials -- 2.2.1 Review of Frequentist Methods and Their Applications in Phase II Clinical Trials -- 2.2.2 Frequentist Methods for Single-Treatment Pilot Studies -- 2.2.3 Frequentist Methods for Comparative Studies -- 2.2.4 Frequentist Methods for Screening Studies -- 2.3 Bayesian Methods in Phase II Clinical Trials -- 2.3.1 Review of Bayesian Methods and Their Application in Phase II Clinical Trials -- 2.3.2 Bayesian Methods for Single-Treatment Pilot Studies, Comparative Studies and Selection Screens -- 2.4 Decision-Theoretic Methods in Phase II Clinical Trials -- 2.5 Analysis of Multiple Endpoints in Phase II Clinical Trials -- 2.6 Outstanding Issues in Phase II Clinical Trials -- 3. Response-Adaptive Designs in Phase III Clinical Trials -- 3.1 Introduction -- 3.2 Adaptive Designs for Binary Treatment Responses -- 3.2.1 Play-the-Winner Design -- 3.2.2 Randomized Play-the-Winner Design -- 3.2.3 Generalized Pólya's Urn (GPU) -- 3.2.4 Randomized Pólya Urn Design -- 3.2.5 Birth-and-Death Urn Design -- 3.2.6 Birth-and-Death Urn with Immigration Design.

3.2.7 Drop-the-Loser Urn Design -- 3.2.8 Sequential Estimation-Adjusted Urn Design -- 3.2.9 Doubly Adaptive Biased Coin Design -- 3.3 Adaptive Designs for Binary Treatment Responses Incorporating Covariates -- 3.3.1 Covariate-Adaptive Randomized Play-the-Winner Design -- 3.3.2 Treatment Effect Mappings -- 3.3.3 Drop-the-Loser Design with Covariate -- 3.4 Adaptive Designs for Categorical Responses -- 3.5 Adaptive Designs for Continuous Responses -- 3.5.1 Nonparametric-Score-Based Allocation Designs -- 3.5.2 Link-Function-Based Allocation Designs -- 3.5.3 Continuous Drop-the-Loser Design -- 3.6 Optimal Adaptive Designs -- 3.7 Delayed Responses in Adaptive Designs -- 3.8 Biased Coin Designs -- 3.9 Real Adaptive Clinical Trials -- 3.10 Data Study for Different Adaptive Schemes -- 3.10.1 Fluoxetine Trial -- 3.10.2 Pregabalin Trial -- 3.10.3 Simulated Trial -- 3.11 Concluding Remarks -- 4. Inverse Sampling for Clinical Trials: A Brief Review of Theory and Practice -- 4.1 Introduction -- 4.1.1 Inverse Binomial Sampling -- 4.1.2 Partial Sequential Sampling -- 4.2 Two-Sample Randomized Inverse Sampling for Clinical Trials -- 4.2.1 Use of Mann-Whitney Statistics -- 4.2.2 Fixed-Width Confidence Interval Estimation -- 4.2.3 Fixed-Width Confidence Interval for Partial Sequential Sampling -- 4.3 An Example of Inverse Sampling: Boston ECMO -- 4.4 Inverse Sampling in Adaptive Designs -- 4.5 Concluding Remarks -- 5. The Design and Analysis Aspects of Cluster Randomized Trials -- 5.1 Introduction: Cluster Randomized Trials -- 5.2 Intracluster Correlation Coefficient and Confidence Interval -- 5.3 Sample Size Calculation for Cluster Randomized Trials -- 5.4 Analysis of Cluster Randomized Trial Data -- 5.5 Concluding Remarks -- PART II EPIDEMIOLOGY -- 6. HIV Dynamics Modeling and Prediction of Clinical Outcomes in AIDS Clinical Research -- 6.1 Introduction.

6.2 HIV Dynamic Model and Treatment Effect Models -- 6.2.1 HIV Dynamic Model -- 6.2.2 Treatment Effect Models -- 6.3 Statistical Methods for Predictions of Clinical Outcomes -- 6.3.1 Bayesian Nonlinear Mixed-Effects Model -- 6.3.2 Predictions Using the Bayesian Mixed-Effects Modeling Approach -- 6.4 Simulation Study -- 6.5 Clinical Data Analysis -- 6.6 Concluding remarks -- 7. Spatial Epidemiology -- 7.1 Space and Disease -- 7.2 Basic Spatial Questions and Related Data -- 7.3 Quantifying Pattern in Point Data -- 7.4 Predicting Spatial Observations -- 7.5 Concluding Remarks -- 8. Modeling Disease Dynamics: Cholera as a Case Study -- 8.1 Introduction -- 8.2 Data Analysis via Population Models -- 8.3 Sequential Monte Carlo -- 8.4 Modeling Cholera -- 8.4.1 Fitting Structural Models to Cholera Data -- 8.5 Concluding Remarks -- 9. Misclassification and Measurement Error Models in Epidemiologic Studies -- 9.1 Introduction -- 9.2 A Few Examples -- 9.2.1 Atom Bomb Survivors Data -- 9.2.2 Coalminers Data -- 9.2.3 Effect of Maternal Dietary Habits on Low Birth Weight in Babies -- 9.3 Binary Regression Models with Two Types of Error -- 9.4 Bivariate Binary Regression Models with Two Types of Error -- 9.5 Models for Analyzing Mixed Misclassified Binary and Continuous Responses -- 9.6 Atom Bomb Data Analysis -- 9.7 Concluding Remarks -- PART III SURVIVAL ANALYSIS -- 10. Semiparametric Maximum-Likelihood Inference in Survival Analysis -- 10.1 Introduction -- 10.2 Examples of Survival Models -- 10.3 Basic Estimation and Limit Theory -- 10.4 The Bootstrap -- 10.4.1 The Regular Case -- 10.4.2 When Slowly Converging Nuisance Parameters are Present -- 10.5 The Profile Sampler -- 10.6 The Piggyback Bootstrap -- 10.7 Other Approaches -- 10.8 Concluding Remarks -- 11. An Overview of the Semi-Competing Risks Problem -- 11.1 Introduction -- 11.2 Nonparametric Inferences.

11.3 Semiparametric One-Sample Inference -- 11.4 Semiparametric Regression Method -- 11.4.1 Functional Regression Modeling -- 11.4.2 A Bivariate Accelerated Lifetime Model -- 11.5 Concluding Remarks -- 12. Tests for Time-Varying Covariate Effects within Aalen's Additive Hazards Model -- 12.1 Introduction -- 12.2 Model Specification and Inferential Procedures -- 12.2.1 A Pseudo-Score Test -- 12.3 Numerical Results -- 12.3.1 Simulation Studies -- 12.3.2 Trace Data -- 12.4 Concluding Remarks -- 12.5 Summary -- Appendix 12A -- 13. Analysis of Outcomes Subject to Induced Dependent Censoring: A Marked Point Process Perspective -- 13.1 Introduction -- 13.2 Induced Dependent Censoring and Associated Identifiability Issues -- 13.3 Marked Point Process -- 13.3.1 Hazard Functions with Marked Point Process -- 13.3.2 Identifiability -- 13.3.3 Nonparametric Estimation -- 13.3.4 Martingales -- 13.4 Modeling Strategy for Testing and Regression -- 13.4.1 Two-Sample Test for Lifetime Utility or Cost -- 13.4.2 Calibration Regression for Lifetime Medical Cost -- 13.4.3 Two-Sample Multistate Accelerated Sojourn-Time Model -- 13.5 Concluding Remarks -- 14. Analysis of Dependence in Multivariate Failure-Time Data -- 14.1 Introduction -- 14.2 Nonparametric Bivariate Survivor Function Estimation -- 14.2.1 Path-Dependent Estimators -- 14.2.2 Inverse Censoring Probability Weighted Estimators -- 14.2.3 NPMLE-Type Estimators -- 14.2.4 Data Application to Danish Twin Data -- 14.3 Non- and Semiparametric Estimation of Dependence Measures -- 14.3.1 Nonparametric Dependence Estimation -- 14.3.2 Semiparametric Dependence Estimation -- 14.3.3 An Application to a Case-Control Family Study of Breast Cancer -- 14.4 Concluding Remarks -- 15. Robust Estimation for Analyzing Recurrent-Event Data in the Presence of Terminal Events -- 15.1 Introduction -- 15.2 Inference Procedures.

15.2.1 Estimation in the Presence of Only Independent Censoring (with All Censoring Variables Observable) -- 15.2.2 Estimation in the Presence of Terminal Events -- 15.3 Large-Sample Properties -- 15.4 Numerical Results -- 15.4.1 Simulation Studies -- 15.4.2 rhDNase Data -- 15.4.3 Bladder Tumor Data -- 15.5 Concluding Remarks -- Appendix 15A -- 16. Tree-Based Methods for Survival Data -- 16.1 Introduction -- 16.2 Review of CART -- 16.3 Trees for Survival Data -- 16.3.1 Methods Based on Measure of Within-Node Homogeneity -- 16.3.2 Methods Based on Between-Node Separation -- 16.3.3 Pruning and Tree Selection -- 16.4 Simulations for Comparison of Different Splitting Methods -- 16.5 Example: Breast Cancer Prognostic Study -- 16.6 Random Forest for Survival Data -- 16.6.1 Breast Cancer Study: Results from Random Forest Analysis -- 16.7 Concluding Remarks -- 17. Bayesian Estimation of the Hazard Function with Randomly Right-Censored Data -- 17.1 Introduction -- 17.1.1 The Random Right-Censorship Model -- 17.1.2 The Bayesian Model -- 17.2 Bayesian Functional Model Using Monotone Wavelet Approximation -- 17.3 Estimation of the Subdensity F* -- 17.4 Simulations -- 17.5 Examples -- 17.6 Concluding Remarks -- Appendix 17A -- PART IV BIOINFORMATICS -- 18. The Effects of Intergene Associations on Statistical Inferences from Microarray Data -- 18.1 Introduction -- 18.2 Intergene Correlation -- 18.3 Differential Expression -- 18.4 Timecourse Experiments -- 18.5 Meta-Analysis -- 18.6 Concluding Remarks -- 19. A Comparison of Methods for Meta-Analysis of Gene Expression Data -- 19.1 Introduction -- 19.2 Background -- 19.2.1 Technology Details and Gene Identification -- 19.2.2 Analysis Methods -- 19.3 Example -- 19.4 Cross-Comparison of Gene Signatures -- 19.5 Best Common Mean Difference Method -- 19.6 Effect Size Method -- 19.7 POE Assimilation Method.

19.8 Comparison of Three Methods.