Comparing the forecasting performances of linear models for electricity prices with high RES penetration

Gianfreda, Angelica; Ravazzolo, Francesco; Rossini, Luca

doi:10.1016/j.ijforecast.2019.11.002

Cited by 49 publications

(30 citation statements)

References 53 publications

(69 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In these models, different autoregressive terms help in predicting future prices and, by determining a suitable choice of lags and time dummies, these models can adapt well to seasonal characteristics. For instance, IJESM 15,1 autoregressive models are widely used as a benchmark against which other models are compared (Misiorek et al, 2006;Uniejewski et al, 2016;Weron and Misiorek, 2008;Gianfreda et al, 2020). In practice, these models can be adapted to handle moving-average terms (Contreras et al, 2003;Cruz et al, 2011;Huurman et al, 2012;Kristiansen, 2012) or external regressors (Contreras et al, 2003;Misiorek et al, 2006;Feuerriegel et al, 2014).…”

Section: Predictive Modelsmentioning

confidence: 99%

Forecasting electricity prices with machine learning: predictor sensitivity

Naumzik

Feuerriegel

2020

IJESM

View full text Add to dashboard Cite

Purpose Trading on electricity markets occurs such that the price settlement takes place before delivery, often day-ahead. In practice, these prices are highly volatile as they largely depend upon a range of variables such as electricity demand and the feed-in from renewable energy sources. Hence, the purpose of this paper is to provide accurate forecasts.. Design/methodology/approach This paper aims at comparing different predictors stemming from supply-side (solar and wind power generation), demand-side, fuel-related and economic influences. For this reason, this paper implements a broad range of non-linear models from machine learning and draw upon the information-fusion-based sensitivity analysis. Findings This study disentangles the respective relevance of each predictor. This study shows that external predictors altogether decrease root mean squared errors by up to 21.96%. A Diebold-Mariano test statistically proves that the forecasting accuracy of the proposed machine learning models is superior. Research limitations/implications The performance gain from including more predictors might be larger than from a better model. Future research should place attention on expanding the data basis in electricity price forecasting. Practical implications When developing pricing models, practitioners can achieve reasonable performance with a simple model (e.g. seasonal-autoregressive moving-average) that is built upon a wide range of predictors. Originality/value The benefit of adding further predictors has only recently received traction; however, little is known about how the individual variables contribute to improving forecasts in machine learning.

show abstract

Section: Predictive Modelsmentioning

confidence: 99%

Forecasting electricity prices with machine learning: predictor sensitivity

Naumzik

Feuerriegel

2020

IJESM

View full text Add to dashboard Cite

show abstract

“…Those uncertainties are reflected in the futures prices as well (Kre car et al, 2019). Many prediction tools useful in the past are not suitable any more to forecast the patterns like spikes and diffusions patterns under the new market circumstances (Gianfreda, Ravazzolo, & Rossini, 2018).…”

Section: The Influence Of Res On the Electricity Marketmentioning

confidence: 99%

Risk mitigation in the electricity market driven by new renewable energy sources

Krecar

Gubina

2019

WIREs Energy & Environment

View full text Add to dashboard Cite

An important indicator of supply and demand uncertainty on electricity markets is risk premium (RP), an integral part of electricity price. The level of RP is describing the risk the market actors expect in the future due to the market uncertainties. With the electricity market price behavior rapidly changing, the rising market supply uncertainty is increasing the volatility of RP. To better understand this behavior, power market participants need new models that will efficiently use the information available in the processes driving the electricity price and RP and to explain the influence of RES uncertainties on the future RP and electricity prices. A decade ago, researchers investigating RP focused primarily on the uncertainties arising from consumption forecast and generation outages. RES share in generation mix was small and its influence on uncertainty was negligible, leading to much lower volatilities of RP than today. With the increasing influence of RES, typically exhibiting variability on a sub-hourly level, traditional models using daily electricity price to calculate RP were becoming inadequate. In this dynamic period of electricity market transformation, this paper highlights the importance of RP signals to market actors. A stochastic method for RP calculation is discussed with the associated RP model, driven by the intraday dynamics. An example of the RP signal is presented on historical price data from the German electricity market, highlighting uncertainty pattern developed over the years. With such an approach, the market actors can adjust their trading strategies thus mitigating their market risk exposure.

show abstract

“…Second, power grids are one of the most critical infrastructures and have a major role in sustainable development and economic growth. The recent innovation in energy production and, in particular, the large increase in renewable energy resources (RES) have added complexity to the management of the electricity system, see Gianfreda et al (2018) for an application of RES to predict day-ahead prices. Moreover, smart grids are the future technologies in power grid development, management, and control, see Yu et al (2011) and Yu and Xue (2016).…”

Section: Introductionmentioning

confidence: 99%

“…In the last years, a large and growing body of literature deals with the forecasting of daily electricity prices (see Weron, 2014, for a review). However, the main focus of the literature is on the forecasting of electricity prices influenced by variables with the same frequency, such as renewable energy sources (Gianfreda et al, 2018) or weather forecasts (Huurman et al, 2012). This empirical application draws on the literature using macroeconomic variables to improve the forecasting performance of single frequency models, due to the fact that macroeconomic variables are of interested in the diagnostic of electricity prices.…”

Section: Introductionmentioning

confidence: 99%

Forecasting Daily Electricity Prices with Monthly Macroeconomic Variables

2019

Self Cite

View full text Add to dashboard Cite

We analyse the importance of macroeconomic information, such as industrial production index and oil price, for forecasting daily electricity prices in two of the main European markets, Germany and Italy. We do that by means of mixed-frequency models, introducing a Bayesian approach to reverse unrestricted MIDAS models (RU-MIDAS). We study the forecasting accuracy for different horizons (from 1 day ahead to 28 days ahead) and by considering different specifications of the models. We find gains around 20% at short horizons and around 10% at long horizons. Therefore, it turns out that the macroeconomic low frequency variables are more important for short horizons than for longer horizons. The benchmark is almost never included in the model confidence set.

show abstract

Comparing the forecasting performances of linear models for electricity prices with high RES penetration

Cited by 49 publications

References 53 publications

Forecasting electricity prices with machine learning: predictor sensitivity

Forecasting electricity prices with machine learning: predictor sensitivity

Risk mitigation in the electricity market driven by new renewable energy sources

Forecasting Daily Electricity Prices with Monthly Macroeconomic Variables

Contact Info

Product

Resources

About