Modeling and Analysis of Longitudinal Data (Volume 50) (Handbook of Statistics, Volume 50)

Longitudinal Data Analysis, Volume 50 in the Handbook of Statistics series covers how data consists of a series of repeated observations of the same subjects over an extended time frame and is thus useful for measuring change. Such studies and the data arise in a variety of fields, such as health sciences, genomic studies, experimental physics, sociology, sports and student enrollment in universities. For example, in health studies, intra-subject correlation of responses must be accounted for, covariates vary with time, and bias can arise if patients drop out of the study. Provides the authority and expertise of leading contributors from an international board of authors Presents the latest release in the Handbook of Statistics series Updated release includes the latest information on Modeling and Analysis of Longitudinal Data Cover image Title page Table of Contents Series Page Copyright Contributors Preface Part I: Foundations Chapter 1 Multivariate and shared parameter mixed-effects models for intensive longitudinal data Abstract 1 Introduction 2 Multivariate mixed-effects models 3 Example 4 Discussion Appendix A SAS PROC NLMIXED code Appendix B STAN code References Chapter 2 Hierarchical and incomplete data Abstract 1 Introduction 2 Case study: Age-related macular degeneration trial 3 Modeling tools for longitudinal data 4 Marginalization and marginal interpretation of hierarchical models 5 Incomplete data 6 Analysis of the ARMD trial 7 Extensions 8 Flexibly accommodating correlation and overdispersion 9 Concluding remarks References Chapter 3 Modeling longitudinal trends in event-related potentials Abstract 1 Introduction 2 An implicit learning paradigm 3 What is an ERP? 4 Preprocessing ERPs 5 Enhancing signal-to-noise ratio for longitudinal modeling 6 Modeling longitudinal trends References Part II: Further methods and models Chapter 4 Longitudinal data analysis by hierarchical state space models Abstract 1 Introduction 2 Linear Gaussian SSMs 3 Hierarchical state space models 4 Estimation, inference and prediction 5 Applications 6 Extensions to bivariate HSSMs 7 Extensions to nonlinear non-Gaussian HSSMs 8 Bibliographical notes References Chapter 5 Latent state-trait analysis Abstract 1 Introduction 2 Background 3 Latent variables in LST theory 4 Consistency, occasion specificity, and reliability 5 The multitrait-multistate (MTMS) model 6 Illustrative application of the MTMS model 7 The Singletrait-Multistate Model with m – 1 Indicator-Specific Factors 8 Illustrative application of the STMS-IS model 9 Discussion Appendix A1: Mplus software code for fitting the MTMS model to the MRT data Appendix A2: Lavaan software code for fitting the MTMS model to the MRT data Appendix B1: Mplus syntax for fitting the STMS-IS model to the MRT data example Appendix B2: Lavaan syntax for fitting the STMS-IS model to the MRT data example References Chapter 6 Recent advances in longitudinal data analysis Abstract 1 Introduction 2 Covariance selection 3 Variable selection 4 Machine learning approaches 5 Discussion References Part III: Applications Chapter 7 Government as population: Demographic perspectives on the United States legislative, judicial and executive branches, 1789–2020 Abstract 1 Introduction 2 Data and sources 3 Population perspectives 4 Applications and projections 5 The political destiny of government demography 6 Conclusions Acknowledgments References Chapter 8 Beginnings: Formation and growth of natural phenomena out of Fisher information Abstract 1 Nature, as evolving, communicating systems: Essential role of background Fisher information 2 What has past use of the MFI principle shown? 3 Information aspects of system coordinates and parameters 4 Nature of the output: Schrodinger's dilemma 5 Concept of system “complexity” 6 Principle of maximum Fisher information (MFI) 7 Definition of Fisher information (FI) I 8 Summary to this point 9 MFI view of the fundamental physical constants per se as statistical entities 10 Discussion 11 Information channels: Basics of MFI principle 12 Application: Early growth laws p(t) for viruses, cancer and the cosmos 13 Regarding the fundamental need for well-defined Planck constants 14 On early growth of cancer, or virus, as quantum-based 15 Growth laws of viruses: Of cancer 16 Experimental verification 17 A sketch of past work on cosmic origins relating to the preceding theory of fluctuations 18 Possible mechanisms for a perpetual input information level J 19 Comparison for early growth of viruses, cancer, the cosmos Appendix A Appendix: The MFI (or EPI) principle is satisfied by power-law solutions for q(t), p(t) Appendix B Appendix: The required exponent in power-law solution p(t) is generally the Fibonacci constant ' References Further reading Index