A hybrid model of kernel density estimation and quantile regression for GEFCom2014 probabilistic load forecasting

Journal article

We present a hybrid model combining two types of probabilistic forecast, a kernel density estimation (KDE) and a quantile regression, as part of the load forecasting track of the Global Energy Forecasting Competition 2014 (GEFCom 2014). The KDE method is initially implemented with a time-decay parameter. We later improve this method by conditioning on the temperature or the period of the week variables to provide more accurate forecasts. Secondly, we develop a simple but effective quantile regression forecast. The novel aspects of our methodology are two-fold. First, we introduce symmetry into the time-decay parameter of the kernel density estimation based forecast and secondly we merge our forecasts with a weighted combination of the three main forecasts with different weights for different periods of the month, which, to our knowledge, has not been applied to electricity load forecasts before.

Authors: Haben, S; Giasemidis, G

• Publication status: Accepted
• Publisher: Elsevier
• Journal: Journal of Forecasting
• Edition: Author's Original
• ISSN: 0169-2070
• Language: English
• Subjects: Probability; Mathematics; Mathematical modeling (engineering)