Risk assessment due to load demand and electricity price forecast uncertainty

  • Gao Gao

Student thesis: Doctoral Thesis


This thesis introduces methodologies for load demand forecasting and electricity price forecasting. The autoregressive integrated moving average (ARIMA) models, seasonal autoregressive integrated moving average (SARIMA) models and artificial neural network (ANN) techniques are introduced to forecast load demand. And the forecasting process of load demand includes monthly, seasonal, annual and multi-step-ahead. Similarly, the same forecasting methods are used for electricity price forecasting. In terms of forecast error analysis method, the root mean square percentage error (RMSPE) and the mean absolute percentage error (MAPE) are used to observe the accuracy of forecasting results. After obtaining the forecasting results, this thesis proposes a risk index method to observe the forecast error more intuitively. The risk indexes are presented based on the load demand forecast errors and electricity price forecast errors respectively. In addition, this thesis investigates the financial risk by combining the errors made by load demand forecasting and electricity price forecasting. The Value-at-Risk (VaR) and Expected Shortfall (ES) methods in economic theory are used to analyse the financial risks. Moreover, to present the actual risk that the market participants have to bear, the daily, monthly, seasonal and annual total financial risks under three different preconditions are compared.
Date of Award1 Oct 2019
Original languageEnglish
Awarding Institution
  • University Of Strathclyde
SponsorsUniversity of Strathclyde
SupervisorKwok Lo (Supervisor) & Ivana Kockar (Supervisor)

Cite this