Handbook of Discrete-Valued Time Series (Chapman & Hall/CRC Handbooks of Modern Statistical Methods)

"Model a Wide Range of Count Time Series Handbook of Discrete-Valued Time Series presents state-of-the-art methods for modeling time series of counts and incorporates frequentist and Bayesian approaches for discrete-valued spatio-temporal data and multivariate data. While the book focuses on time series of counts, some of the techniques discussed can be applied to other types of discrete-valued time series, such as binary-valued or categorical time series.Explore a Balanced Treatment of Frequentist and Bayesian Perspectives Accessible to graduate-level students who have taken an elementary class in statistical time series analysis, the book begins with the history and current methods for modeling and analyzing univariate count series. It next discusses diagnostics and applications before proceeding to binary and categorical time series. The book then provides a guide to modern methods for discrete-valued spatio-temporal data, illustrating how far modern applications have evolved from their roots. The book ends with a focus on multivariate and long-memory count series.Get Guidance from Masters in the FieldWritten by a cohesive group of distinguished contributors, this handbook provides a unified account of the diverse techniques available for observation- and parameter-driven models. It covers likelihood and approximate likelihood methods, estimating equations, simulation methods, and a Bayesian approach for model fitting."--Provided by publisher. Read more... Abstract: "Model a Wide Range of Count Time Series Handbook of Discrete-Valued Time Series presents state-of-the-art methods for modeling time series of counts and incorporates frequentist and Bayesian approaches for discrete-valued spatio-temporal data and multivariate data. While the book focuses on time series of counts, some of the techniques discussed can be applied to other types of discrete-valued time series, such as binary-valued or categorical time series.Explore a Balanced Treatment of Frequentist and Bayesian Perspectives Accessible to graduate-level students who have taken an elementary class in statistical time series analysis, the book begins with the history and current methods for modeling and analyzing univariate count series. It next discusses diagnostics and applications before proceeding to binary and categorical time series. The book then provides a guide to modern methods for discrete-valued spatio-temporal data, illustrating how far modern applications have evolved from their roots. The book ends with a focus on multivariate and long-memory count series.Get Guidance from Masters in the FieldWritten by a cohesive group of distinguished contributors, this handbook provides a unified account of the diverse techniques available for observation- and parameter-driven models. It covers likelihood and approximate likelihood methods, estimating equations, simulation methods, and a Bayesian approach for model fitting."--Provided by publisher Content: Front Cover Contents Preface Editors Contributors Section I: Methods for Univariate Count Processes Chapter 1: Statistical Analysis of Count Time Series Models: A GLM Perspective Chapter 2: Markov Models for Count Time Series Chapter 3: Generalized Linear Autoregressive Moving Average Models Chapter 4: Count Time Series with Observation-Driven Autoregressive Parameter Dynamics Chapter 5: Renewal-Based Count Time Series Chapter 6: State Space Models for Count Time Series Chapter 7: Estimating Equation Approaches for Integer-Valued Time Series Models. Chapter 8: Dynamic Bayesian Models for Discrete-Valued Time SeriesSection II: Diagnostics and Applications Chapter 9: Model Validation and Diagnostics Chapter 10: Detection of Change Points in Discrete-Valued Time Series Chapter 11: Bayesian Modeling of Time Series of Counts with Business Applications Section III: Binary and Categorical-Valued Time Series Chapter 12: Hidden Markov Models for Discrete-Valued Time Series Chapter 13: Spectral Analysis of Qualitative Time Series Chapter 14: Coherence Consideration in Binary Time Series Analysis. Section IV: Discrete-Valued Spatio-Temporal ProcessesChapter 15: Hierarchical Dynamic Generalized Linear Mixed Models for Discrete-Valued Spatio-Temporal Data Chapter 16: Hierarchical Agent-Based Spatio-Temporal Dynamic Models for Discrete-Valued Data Chapter 17: Autologistic Regression Models for Spatio-Temporal Binary Data Chapter 18: Spatio-Temporal Modeling for Small Area Health Analysis Section V: Multivariate and Long Memory Discrete-Valued Processes Chapter 19: Models for Multivariate Count Time Series. Chapter 20: Dynamic Models for Time Series of Counts with a Marketing ApplicationChapter 21: Long Memory Discrete-Valued Time Series Back Cover. "Model a Wide Range of Count Time Series Handbook of Discrete-Valued Time Series presents state-of-the-art methods for modeling time series of counts and incorporates frequentist and Bayesian approaches for discrete-valued spatio-temporal data and multivariate data. While the book focuses on time series of counts, some of the techniques discussed can be applied to other types of discrete-valued time series, such as binary-valued or categorical time series. Explore a Balanced Treatment of Frequentist and Bayesian Perspectives Accessible to graduate-level students who have taken an elementary class in statistical time series analysis, the book begins with the history and current methods for modeling and analyzing univariate count series. It next discusses diagnostics and applications before proceeding to binary and categorical time series. The book then provides a guide to modern methods for discrete-valued spatio-temporal data, illustrating how far modern applications have evolved from their roots. The book ends with a focus on multivariate and long-memory count series. Get Guidance from Masters in the FieldWritten by a cohesive group of distinguished contributors, this handbook provides a unified account of the diverse techniques available for observation- and parameter-driven models. It covers likelihood and approximate likelihood methods, estimating equations, simulation methods, and a Bayesian approach for model fitting."--Provided by publisher