Modelling and Forecasting Energy Time Series Using Stata


Timberlake Consultants

Start Date:

End Date:

Application Deadline:

Type

Professional training

Language of Instruction

English

Certifications & Titles

Certificate of attendance

Study Options

Full Time

Funding Options

Students are able to apply for a 50% discount off the listed course fees - see website for details

Location

Cass Business School, London

London

United Kingdom

Start Date:

End Date:

Location

United Kingdom

London

Type

Professional training

Application Deadline:


Over the past two decades several countries around the world have liberalized their energy markets, with the result that the risk for utilities, energy producers and commodity market operators has increased substantially.

In this new environment, accurate modelling and forecasting of energy demand and prices has become of key importance for all market players to plan their short- and long-term operations.

This course provides a review of and a practical guide to several major econometric methodologies to modelling the stylised facts of the energy prices and demand time series, via regression and cointegration analysis, univariate and multivariate GARCH models. Practical demonstrations will be conducted using Stata.

AGENDA

Day 1: Modelling the conditional mean of energy demands and prices series

  • Introduction to energy time series features: distributional properties, stationarity, seasonality, autocorrelation, heteroscedasticity.
  • Univariate models of conditional mean (MA, AR, ARMA, ARIMA, ARMAX). Analysis of the properties and practical applications of identification and diagnostic checking of ARMA models.
  • Forecasting energy prices with ARMA models.
  • Vector Autoregressive (VAR) models. Analysis of the properties and practical applications of identification and diagnostic checking of VAR models.
  • Long-run relationships in energy: applying cointegration analysis to model fuels demand.

Day 2: Modelling the volatility of energy prices series

  • Characteristics of energy prices volatility.
  • ARCH and GARCH models, Integrated GARCH model, GARCH in mean. Asymmetric GARCH models: SAARCH, EGARCH, GJR, TGARCH, APARCH. Estimating the news impact curve.
  • Forecasting energy prices volatility with GARCH models
  • Multivariate GARCH models: Diagonal VECH, Constant Conditional Correlation, Dynamic Conditional Correlation models. Analysis of the properties and practical applications of identification and diagnostic checking of MGARCH models.
  • Forecasting energy markets correlations with MGARCH models.