A Course In Time Series Analysis Daniel Pena George C Tiao Ruey S Tsay by Daniel Peña; George C Tiao; Ruey S Tsay 9780471361640, 047136164X instant download after payment.

This book is based on the lectures of the ECAS' 97 Course in Time Series Analysis

held at El Escorial, Madrid, Spain, from September 15 to September 19, 1997. The

course was sponsored by the European Courses in Advanced Statistics (ECAS). In

accordance with the objectives of ECAS, the lectures are directed to both researchers

and teachers of statistics in academic institutions and statistical professionals in industry

and govermment, with the goal of presenting an overview of the current status

of the area. In particular, different approaches to time series analysis are discussed and

compared. In editing the book, we have worked hard to uphold ECAS' objectives. In

addition, special efforts have been made to unify the notation and to include as many

topics as possible, so that readers of the book can have an overview of the current

status of time series research and applications.

The book consists of three main components. The first component concern basic

materials of univariate time series analysis presented in the first eight chapters. It

includes recent developments in outlier detection, automatic model selection, and

seasonal adjustment. The second component addresses advanced topics in univariate

time series analysis such as conditional heteroscedastic models, nonlinear models,

Bayesian analysis, nonparametric methods, and neural networks. This component

represents current research activities in univariate time series analysis. The third and

final component of the book concerns with multivariate time series, including vector

ARMA models, cointegration, and linear systems.

The book can be used as a principal text or a complementary text for courses

in time series. A basic time series course can be taught from the first part of the

book that presents the basic material that can be found in the standard texts in time

series. This part also includes topics not normally covered in these texts, such as

the extended and inverse autocorrelation function, the decomposition of the forecast

function of ARIMA models, a detailed analysis of outliers and influential observations

and automatic methods for model building and model based seasonal adjustment. For

a basic course this book should be complemented with some of the excellent texts

available. The book would be very well suited for an advanced course in which

some of the basic material can be quickly reviewed using the first part, that skips

many details and concentrates in the main concepts of general applicability. Then the course can concentrate in the topics in Parts 2 and 3. If the scope of the course is more

in methodological extensions of univariate linear models the material in Part 2 can

be useful, whereas if the objective is to introduce multivariate modeling Part 3 will

be appropriate. To facilitate the use of the book as a text, all the time series data used

in this book can be down loaded from the web address: http://gsbwww.

uchicago. edu/fac/ruey. tsay/teaching/ecas/