Time Series Analysis

The course is taught to PhD & Masters students and considers the use of modern time series methods. Topics covered include an introduction to the dynamic properties of time series, stochastic difference equations, stationary univariate models, forecast evaluation, state-space models, non-stationary models and unit roots, vector autoregression models, structural vector autoregression models, Bayesian vector autoregression models, cointegration and error-correction, dynamic factor models and factor augmented vector autoregression models, heteroskedastic and stochastic volatility models, as well as nonlinear regime-switching models. Throughout the course we will emphasize areas of ongoing research.

1) Introduction [notes] [slides] [tutorial] [R files]

2) ARMA models [notes] [slides] [tutorial] [R files]

3) Forecasting [notes] [slides] [tutorial] [R files]

4) State-space models [notes] [slides] [tutorial] [R files]

5) Decomposing time series data [notes] [slides] [tutorial] [R files]

6) Nonstationarity [notes] [slides] [tutorial] [R files]

7) Vector autoregression models [notes] [slides] [tutorial] [R files]

8) Structural vector autoregression models [notes] [slides] [tutorial] [R files]

9) Bayesian vector autoregression models [notes] [slides] [tutorial] [R files]

10) Cointegration [notes] [slides] [tutorial] [R files]

11) Volatility models [notes] [slides] [tutorial] [R files]

12) Nonlinear models [notes] [slides] [tutorial] [R files]

Use Chrome to print the HTML documents. You may wish to scale the document to 85% when doing so.