In this chapter, we consider some advanced time series methods and their implementation using R. Time series analysis, as a discipline, is broad enough to fill hundreds of books (the most important references, both in theory and R programming, will be listed at the end of this chapter's reading list); hence, the scope of this chapter is necessarily highly selective, and we focus on topics that are inevitably important in empirical finance and quantitative trading. It should be emphasized at the beginning, however, that this chapter only sets the stage for further studies in time series analysis.

Our previous book Introduction to R for Quantitative Finance, Packt Publishing, discusses some fundamental topics of time series analysis such as linear, univariate time series modeling, Autoregressive integrated moving average (ARIMA), and volatility modeling Generalized Autoregressive Conditional Heteroskedasticity (GARCH). If you have never worked with R for time series analysis, you might want to consider going through Chapter 1, Time Series Analysis of that book as well.

The current edition goes further in all of these topics and you will become familiar with some important concepts such as cointegration, vector autoregressive models, impulse-response functions, volatility modeling with asymmetric GARCH models including exponential GARCH and Threshold GARCH models, and news impact curves. We first introduce the relevant theories, then provide some practical insights to multivariate time series modeling, and describe several useful R packages and functionalities. In addition, using simple and illustrative examples, we give a step-by-step introduction to the usage of R programming language for empirical analysis.