The book with the title ‘Bayesian Analysis of Time Series’ by Lyle D. Broemeling is an excellent source to learn time series concepts, methods, expressions, and interpretations from the Bayesian viewpoint using R code and WinBugs code. Graduate-level background in statistical theory and software skill is necessary to read and comprehend the contents of this book. Noteworthy features in this book are several examples from biology, agriculture, business, economics, sociology, and astronomy, exercises in every chapter. Examples include airline passenger booking, sunspots, Los Angeles rainfall. The book is suitable for usage to teach in a graduate-level Bayesian time series course. There are 10 well-written chapters covering the topics introduction to Bayesian concepts especially in time series, Monte Carlo Markov Chain, Metropolis algorithm, Gibbs sampling, preliminary considerations for time series analysis, basic random models, time series versus regression, stationarity, spectral analysis, dynamic linear models, shift point problem, residuals versus diagnostic tests, predictive distributions, Kalman filter, adaptive estimation among others. The references are exhaustive and well selected for the readers. The exercises are challenging.
The book with the title ‘Analysis of Time Series: An Introduction with R’ by Chris Chatfield and Haipeng Xing presents a balanced and comprehensive illustration of time series theory and practices. Noteworthy features in this book are, Fourier, Laplace, and z transforms, Dirac delta function, answers to the challenging exercises in every chapter. Basic knowledge of statistical theory and software skill is necessary to read and comprehend the contents of the book. There are 14 well-written chapters and three appendices covering the topics introduction, descriptive techniques, linear time series, time-domain approach, ARIMA models, forecasting methods, spectral analysis, linear systems, state-space models, Kalman filter, nonlinear models, volatility models, bivariate processes, multivariate models, advanced topics among others. The references are exhaustive and well selected for the readers.
The book with the title ‘Time Series: A Data Analysis Approach Using R’ by Robert H. Shumway and David S. Stoffer is an excellent source for versatile time-series analytic methods for dependent data. This book exposes both the time domain and frequency domain approaches. A good background in statistical theory is necessary to read and comprehend the materials in the book. Noteworthy examples include climate changes, pain perceptions, magnetic image resonance, economic, and financial problems. There are eleven well-written chapters and four appendices covering a wide range of topics time series basic elements, correlation, dependence, stationarity, least squares, exploratory techniques, smoothing, ARIMA, spectral analysis, GARCH models, unit root testing, memory, fractional differencing, state-space models, prewhitening, bootstrapping, notes for using R codes, complex primer, maximum likelihood, causality, invertibility, ARCH model theory, probability and statistics primer among others. The problems in every chapter are quite challenging. The book is suitable for usage to teach a graduate-level time series. The references are limited and core for the readers.
I enjoyed reading these three books. I recommend highly these three books to statistics and computing professionals.