Statistical methods for modeling human dynamics : an interdisciplinary dialogue / edited by Sy-Miin Chow, Emilio Ferrer, Fushing Hsieh.

Format:

Book

Contributor:

Chow, Sy-Miin.

Ferrer, Emilio.

Hsieh, Fushing.

Series:

Notre Dame series on quantitative methodologies

The Notre Dame series on quantitative methodology

Language:

English

Subjects (All):

Human behavior--Mathematical models.

Human behavior.

Dyadic analysis (Social sciences).

Psychometrics.

Sociometry.

Physical Description:

xvi, 428 pages : illustrations ; 24 cm.

Place of Publication:

New York : Routledge, [2010]

Summary:

This interdisciplinary volume features contributions from researchers in the fields of psychology, neuroscience, statics, computer science, and physics. State-of-the-art techniques and applications used to analyze data obtained from studies in cogniton, emotion, and electrophysiology are reviewed along with techniques for modeling in real time and for examining lifespan cognitive changes, for conceptualizing change using item response, nonparametric and heirarchical models, and control theory-inspired techniques for deriving diagnoses in medical and psychotherapeutic settings. The syntax for running the analyses presented in the book is provided at www.psypress.com/chow. Most of the programs are written in R while others are for MATLABʼ, SAS, Win-BUGS, and DyFA.

Readers will appreciate a review of the latest methodological techniques developed in the last few years. Highlights include an examination of:

Statistical and mathematical modeling techniques for the analysis of brain imaging such as EEGs, fMRIs, and other neuroscience data

Dynamic modeling techniques for intensive repeated measurement data

Panel modeling techniques for fewer time points data

State-space modeling techniques for psychological data

Techniques used to analyze reaction time data

Each chapter features an introductory overview of the techniques needed to understand the chapter, a summary, and numerous examples. Each self-contained chapter can be read on its own and in any order. Divided into three major sections, the book examines techniques for examining within-person derivatios in change patterns, intra-individual change, and inter-individual differences in change and interpersonal dynamics. Intended for advanced students and researchers, this book will appeal to those intersted in applying state-of-the-art dynamic modeling techniques to the study of neuroscience, developmental, cognitive, and social/personality psychology, as well as neuroscience, computer science, and engineering.

Contents:

Chapter 1 Introduction and Section Overview 1

1.1 Part I: Parametric and Exploratory Approaches for Extracting Wimin-Person Nonstationarities 1

1.2 Part II: Representing and Extracting Intraindividual Change 4

1.3 Part III: Modeling Interindividual Differences in Change and Interpersonal Dynamics 6

References 8

Part I Parametric and Exploratory Approaches for Extracting Within-Person Nonstationarities

Chapter 2 Dynamic Modeling and Optimal Control of Intraindividual Variation: A Computational Paradigm for Nonergodic Psychological Processes 13

2.1 Introduction 13

2.2 Ergodicity 14

2.3 (Lack of) Homogeneity 16

2.4 Nonstationarity 20

2.5 Illustrative EKFIS Application to a Nonstationary Time Series 24

2.6 A Monte Carlo Study 27

2.7 Optimal Control 31

2.8 Conclusion 35

References 35

Chapter 3 Dynamic Spectral Analysis of Biomedical Signals with Application to Electroencephalogram and Heart Rate Variabilityp 39

3.1 Introduction 39

3.2 Biomedical Signals 41

3.3 Time-Frequency Representations 50

3.4 Parametric Time-Varying Spectrum Estimation 54

3.5 Case Study I: Estimation of ERS of EEG 68

3.6 Case Study II: Estimation of HRV Dynamics During an Orthostatic Test 72

3.7 Discussion 78

Acknowledgments 80

References 80

Chapter 4 Cluster Analysis for Nonstationary Time Series 85

4.1 Introduction 85

4.2 Fourier Analysis 89

4.3 The WP Transform 92

4.4 Clustering Nonstationary Time Series 98

4.5 Simulations 103

4.6 Illustrative Example 109

4.7 Summary 112

Acknowledgments 113

Appendix 4.1 Estimation of the Posterior Probability in Equation 4.4 114

Appendix 4.2 BBA for Selecting the Best Clustering Basis 115

Appendix 4.3 Model-Based Feature Selection Algorithm 117

References 120

Chapter 5 Characterizing Latent Structure in Brain Signals 123

5.1 Introduction 123

5.2 Inferring Latent Structure via AR and TVAR Models 127

5.3 Detecting Fatigue from EEGs: Experimental Setting and Data Analysis 137

5.4 Conclusions and Future Directions 150

Acknowledgments 152

Appendix 5.1 Posterior Estimation in NDLMs 152

References 153

Chapter 6 A Closer Look at Two Approaches for Analysis and Classification of Nonstationary Time Series 155

Part II Representing and Extracting Intraindividual Change

Chapter 7 Generalized Local Linear Approximation of Derivatives from Time Series 161

7.1 Introduction 161

7.2 Time Delay Embedding 163

7.3 LLA Estimates of Derivatives 165

7.4 LDE Estimates of Derivatives 165

7.5 Relationship between LLA and the LDE Loading Matrix 167

7.6 Simulation 169

7.7 Example Application 171

7.8 Example Program 173

7.9 Modeling Results 174

7.10 Discussion 176

7.11 Conclusions 176

Acknowledgments 177

References 177

Chapter 8 Unbiased, Smoothing-Corrected Estimation of Oscillators in Psychology 179

8.1 How do Individuals Change over Time? When? Why? 179

8.2 Method for x-Corrected Estimation of Parameters 192

8.3 Estimation of co and p 194

8.4 Nonoscillating Time Series 201

8.5 Conclusions 207

Appendix 8.1 209

References 210

Chapter 9 Detrendrng Response Time Series 213

9.1 Introduction 213

9.2 Motivating Series 217

9.3 Defending Methods 220

9.4 A Simulation Study 229

9.5 Discussion and Conclusions 237

Acknowledgments 238

References 239

Chapter 10 Dynamic Factor Analysis with Ordinal Manifest Variables 241

10.1 Introduction 241

10.2 DFA Models and their Estimation 243

10.3 Polychoric Lagged Correlations 246

10.4 A Simulation Study 249

10.5 An Empirical Example 255

10.6 Concluding Comments 260

Acknowledgments 262

References 262

Chapter 11 Measuring Intraindividual Variability with Intratask Change Item Response Models 265

11.1 Introduction 265

11.2 Intratask Change Item Response Models 269

11.3 Simulations 275

11.4 Example: IIV and Working Memory 277

11.5 Discussion 279

Acknowledgments 283

References 283

Part III Modeling Interindividual Differences in Change and Interpersonal Dynamics

Chapter 12 Developing a Random Coefficient Model for Nonlinear Repeated Measures Data 289

12.1 Introduction 289

12.2 Alternative Models for the MNREAD Data 295

12.3 A Random Coefficient Model for the MNREAD Data 310

12.4 Discussion 315

Appendix 12.1 The Quadratic-Linear Model with a Smooth Transition between Phases 316

References 317

Chapter 13 Bayesian Discrete Dynamic System by Latent Difference Score Structural Equations Models for Multivariate Repeated Measures Data 319

13.1 Be Methods 321

13.2 Part I: Fitting a Univariate Latent Difference Score Model 324

13.3 Part II: Fitting a Bivariate Difference Score Model 330

13.4 Discussion 339

References 345

Chapter 14 Longitudinal Mediation Analysis of Training Intervention Effects 349

14.1 Introduction 349

14.2 Mediation Analysis 350

14.3 Methods for the Analysis of Training Intervention with Mediation Effects 353

14.4 Empirical Data Analysis 362

14.5 Conclusion and Discussion 376

References 378

Chapter 15 Exploring Intraindividual, Interindividual, and Intervariable Dynamics in Dyadic Interactions 381

15.1 Introduction: Dyadic Interactions 381

15.2 Illustrative Data: Daily Fluctuations in Affect 384

15.3 Lempell-Ziv (L-Z) Complexity 384

15.4 Hierarchical Segmentation 389

15.5 Stochastic Transition Networks 398

15.6 Discussion 407

Acknowledgment 409

References 409.

Notes:

Includes bibliographical references and indexes.

ISBN:

9781848728257

1848728255

OCLC:

458737715

Publisher Number:

99939979535