Search results

Empirical publications on the time series modeling and forecasting of electricity prices vary widely regarding the conditions, and the findings make it difficult to generalize results. Against this background, it is surprising that there is a lack of statistics-based literature reviews on the forecasting performance when comparing different models. The purpose of the present study is to fill this gap.

The authors conduct a comprehensive literature analysis from 2000 to 2015, covering 86 empirical studies on the time series modeling and forecasting of electricity spot prices. Various statistics are presented to characterize the empirical literature on electricity spot price modeling, and the forecasting performance of several model types and modifications is analyzed. The key issue of this study is to offer a comparison between different model types and modeling conditions regarding their forecasting performance, which is referred to as a quasi-meta-analysis, i.e. the analysis of analyses to achieve more general findings independent of the circumstances of single studies.

The authors find evidence that generalized autoregressive conditional heteroscedasticity models outperform their autoregressive–moving-average counterparts and that the consideration of explanatory variables improves forecasts.

To the best knowledge of the authors, this paper is the first to apply the methodology of meta-analyses in a literature review of the empirical forecasting literature on electricity spot markets.

Several research papers related to electricity price forecasting have been reported in the leading journals in last 20 years. The purpose of this paper is to present a comprehensive survey and comparison of these techniques.

The present article provides an overview of the statistical short‐term price forecasting (STPF) models. The basic theory of these models, their further classification and their suitability to STPF has been discussed. Quantitative evaluation of the performance of these models in the framework of accuracy achieved and computation time taken has been performed. Some important observations of the literature survey and key issues regarding STPF methodologies are analyzed.

It has been observed that price forecasting accuracy of the reported models in day‐ahead markets is better as compared to that in real time markets. From a comparative analysis perspective, there is no hard evidence of out‐performance of one model over all other models on a consistent basis for a very long period. In some of the studies, linear models like dynamic regression and transfer function have shown superior performance as compared to non‐linear models like artificial neural networks (ANNs). On the other hand, recent variations in ANNs by employing wavelet transformation, fuzzy logic and genetic algorithm have shown considerable improvement in forecasting accuracy. However more complex models need further comparative analysis.

This paper is intended to supplement the recent survey papers, in which the researchers have restricted the scope to a bibliographical survey. Whereas, in this work, after providing detailed classification and chronological evolution of the STPF techniques, a comparative summary of various price‐forecasting techniques, across different electricity markets, is presented.

Price forecasting is essential for risk management in deregulated electricity markets. The purpose of this paper is to propose a hybrid technique using wavelet transform (WT) and multiple linear regression (MLR) to forecast price profile in electricity markets.

Price series is highly volatile and non‐stationary in nature. In this work, initially complete price series has been decomposed into separate 48 half‐hourly series and then these series have been categorized into different segments for price forecasting. For some segments, WT based MLR has been applied and for the other segments, simple MLR model has been applied. The model is general in nature and has been implemented for one day‐ahead price forecasting in National Electricity Market (NEM) of Australia. Participants can use the technique practically, since it predicts price well before submission of bids.

Forecasting performance of the proposed WT and MLR based mixed model has been compared with the three other models, an analytical model, a MLR model and an artificial neural network (ANN) based model. The proposed model was found to be better. Performance evaluation for different wavelets was performed, and it has been observed that for improving forecasting accuracy using WT, Daubechies wavelet of order two gives the best performance.

Forecasting accuracy improvement of an established technique by incorporating time domain and wavelet domain variables of the same time series into one set has been implemented in this work. The paper also attempts to explain how non‐stationarity can be removed from a non‐stationary time series by applying WT after appropriate statistical investigation. Moreover, real time electricity markets are highly unpredictable and yet under investigated. The model has been applied to NEM for the same reason.

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..

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.

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.

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.

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.

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.

Study conditions of empirical publications on time series modeling and forecasting of electricity prices vary widely, making it difficult to generalize results. The key purpose of the present study is to offer a comparison of different model types and modeling conditions regarding their forecasting performance.

The authors analyze the forecasting performance of AR (autoregressive), MA (moving average), ARMA (autoregressive moving average) and GARCH (generalized autoregressive moving average) models with and without the explanatory variables, that is, power consumption and power generation from wind and solar. Additionally, the authors vary the detailed model specifications (choice of lag-terms) and transformations (using differenced time series or log-prices) of data and, thereby, obtain individual results from various perspectives. All analyses are conducted on rolling calibrating and testing time horizons between 2010 and 2014 on the German/Austrian electricity spot market.

The main result is that the best forecasts are generated by ARMAX models after spike preprocessing and differencing the data.

The present study extends the existing literature on electricity price forecasting by conducting a comprehensive analysis of the forecasting performance of different time series models under varying market conditions. The results of this study, in general, support the decision-making of electricity spot price modelers or forecasting tools regarding the choice of data transformation, segmentation and the specific model selection.

In short-term forecasting of day-ahead electricity prices, incorporating intraday dependencies is vital for accurate predictions. However, it quickly leads to dimensionality problems, i.e. ill-defined models with too many parameters, which require an adequate remedy. This study addresses this issue.

In an application for the German/Austrian market, this study derives variable importance scores from a random forest algorithm, feeds the identified variables into a support vector machine and compares the resulting forecasting technique to other approaches (such as dynamic factor models, penalized regressions or Bayesian shrinkage) that are commonly used to resolve dimensionality problems.

This study develops full importance profiles stating which hours of which past days have the highest predictive power for specific hours in the future. Using the profile information in the forecasting setup leads to very promising results compared to the alternatives. Furthermore, the importance profiles provide a possible explanation why some forecasting methods are more accurate for certain hours of the day than others. They also help to explain why simple forecast combination schemes tend to outperform the full battery of models considered in the comprehensive comparative study.

With the information contained in the variable importance scores and the results of the extensive model comparison, this study essentially provides guidelines for variable and model selection in future electricity market research.

To combine the forecasting by single method using influence information fully, other than regular combined methods only focusing on historical forecasting errors.

To combine the single methods based on the analysis of improved gray correlation, with more related information being considered to enhance the price forecasting precision, such as the trend of the prices, the historical forecasting errors, and the temporal influence factors on prices.

A case of PJM market of USA shows that the proposed method has better performance than any other combined methods, and all single models as well.

The combined performance depends on the forecasting precision of single methods, and the correlation between the single methods, as well as the number of single method that to be combined.

It is a novel idea for combined method to forecasting the time series data, such as electricity prices, electric power loads.

The proposed method considers all the following factors: the similarity between the trends of the single forecasting, the errors of the single models and the temporal influence.

Following the deregulation of electricity markets in the USA, independent power producers operate as for-profit entities. Their profit depends on the price of electricity and an accurate forecast is critical in making bidding decisions on the electricity and reserve markets or engaging in bilateral contracts. Competing price forecasts have their accuracy expressed in statistical terms but producers need to determine the long-term value of using a given forecast. The purpose of this paper is to address this issue by presenting a method of electricity price forecast valuation which compares forecast models using financial rather than statistical measures.

The objectives of this paper are achieved by mathematical modeling of thermal power plants and price forecast information available to market participants and simulating the operation of a thermal power plant using various price forecasts and perfect information (as a baseline). The operating profit calculated over a long period was used for ranking forecast models.

The framework can be used to estimate the value of a new price forecast as well as to determine if potential gains from developing or acquiring a new forecast will justify the expenses. The results show that an improvement in terms of statistical forecast accuracy measures does not guarantee increased profit.

This paper presents a new method for comparing electricity price forecast models. It can be adapted to various types of thermal power plants that operate on liberalized electricity markets and utilize price-based dynamic economic dispatch models.

This paper presents a simulation-based valuation framework for short-term electricity price. The approach described in this paper can be utilized by independent power producers for different types of generators, operating on deregulated electricity markets.

This paper set out to use an autoregressive conditional heteroscedasticity (ARCH)‐type model to capture the time‐varying conditional variance of Alberta electricity prices. This is of major importance in forecasting, since ARCH‐type models allow the conditional variance to depend on elements of the information set.

The paper uses the model to perform static and dynamic forecasts over different horizons and to compare its forecasting performance with a random walk and a moving average model.

The paper provides a study of hourly electricity prices using recent advances in the financial econometrics literature.

The contribution of the paper is its use of models of changing volatility to properly identify the type of heteroscedasticity in the data‐generation processes. This is of major importance in forecasting.

This paper aims to present the results of the comparison between different approaches to the prediction of electricity prices. It is well-known that the properties of the data generation process may prefer some modeling methods over the others. The data having an origin in social or market processes are characterized by unexpectedly wide realization space resulting in the existence of the long tails in the probabilistic density function. These data may not be easy in time series prediction using standard approaches based on the normal distribution assumptions. The electricity prices on the deregulated market fall into this category.

The paper presents alternative approaches, i.e. memory-based prediction and fractal approach compared with established nonlinear method of neural networks. The appropriate interpretation of results is supported with the statistical data analysis and data conditioning. These algorithms have been applied to the problem of the energy price prediction on the deregulated electricity market with data from Polish and Austrian energy stock exchanges.

The first outcome of the analysis is that there are several situations in the task of time series prediction, when standard modeling approach based on the assumption that each change is independent of the last following random Gaussian bell pattern may not be a true. In this paper, such a case was considered: price data from energy markets. Electricity prices data are biased by the human nature. It is shown that more relevant for data properties was Cauchy probabilistic distribution. Results have shown that alternative approaches may be used and prediction for both data memory-based approach resulted in the best performance.

“Personalization” of the model is crucial aspect in the whole methodology. All available knowledge should be used on the forecasted phenomenon and incorporate it into the model. In case of the memory-based modeling, it is a specific design of the history searching routine that uses the understanding of the process features. Importance should shift toward methodology structure design and algorithm customization and then to parameter estimation. Such modeling approach may be more descriptive for the user enabling understanding of the process and further iterative improvement in a continuous striving for perfection.

Memory-based modeling can be practically applied. These models have large potential that is worth to be exploited. One disadvantage of this modeling approach is large calculation effort connected with a need of constant evaluation of large data sets. It was shown that a graphics processing unit (GPU) approach through parallel calculation on the graphical cards can improve it dramatically.

The modeling of the electricity prices has big impact of the daily operation of the electricity traders and distributors. From one side, appropriate modeling can improve performance mitigating risks associated with the process. Thus, the end users should receive higher quality of services ultimately with lower prices and minimized risk of the energy loss incidents.

The use of the alternative approaches, such as memory-based reasoning or fractals, is very rare in the field of the electricity price forecasting. Thus, it gives a new impact for further research enabling development of better solutions incorporating all available process knowledge and customized hybrid algorithms.