This online course using EViews provides a complete introduction to modelling and forecasting volatility models in EViews. The course provides a sound and practical understanding of the GARCH model and its more advanced extensions such as the EGARCH, the TARCH, the APARCH and the IGARCH models used in time series and financial applications, together with a strong background in forecasting. The course targets researchers, practitioners and policy makers who are interested in gaining an in-depth knowledge on how to model volatility in time series and how to use the GARCH class of models in their current or future assignments.
The course runs from 13.00 - 16.00 (UK Time) / 17:00 - 20:00 (UAE Time)
Linear regression is one of the primary tools for econometric and statistical analysis. There is however, considerable evidence that it is not appropriate in modelling and forecasting volatilities. This course provides a comprehensive introduction to GARCH univariate and multivariate models in EViews' non-linear models that successfully predict volatilities and correlations of time series. Taking a “learning-by-doing” approach we aim to present the GARCH model and its extensions: the EGARCH, the APARCH, the TARCH and the IGARCH models. Multivariate extensions are the GARCH are discussed with specific reference to the CCC, DCC and the BEKK. The course employs plenty of financial and macoreconomic data and a constant stream of challenging exercises.
The course specifically focuses on forecasting methodologies with volatility models. Participants will leave iwth the know-how on a wide range of volatility models and the ability to identify which one to use for a specific forecasting purpose.
Advanced topics such as regime switching and Fractionally integrated GARCH can be discussed according to participants’ needs and background.
The course is intentionally flexible and the agenda emerges dynamically depending on the group’s prior background and knowledge of EViews. By the end of the course, all participants will feel comfortable with the following: