Disaggregating time series on multiple criteria for robust forecasting: The case of long-term electricity demand in Greece

摘要: Abstract Electricity demand forecasting is an essential process in the operation and planning procedures of power systems that considerably influences decisions utility providers. Main aim this paper is, first, to examine relationship between a time series influential multiple criteria, and, second, provide long-term electricity forecasts Greece. An original disaggregation or ordinal regression analysis methodological framework outlined optimally assess robust additive value model which as consistent possible with given series. The accuracy stability modeling approach guaranteed through calculation statistical error measures robustness indices, respectively. For case Greece, inferred from data related training period 1999–2013. proposed method has been applied for annual total net Greek interconnected system during following testing 2014–2016. implies level economic growth, represented by national gross domestic product, imposes greatest influence on followed energy efficiency progress weather conditions country. models perform better than linear (least-squares) model, terms prediction reliability, resulting into minimum MAPE equal 0.74%. exact also extraction projections till 2027 based alternative growth scenarios, indicating constant increase