Search

Begin New Search
Proceed to Checkout

Search Results for All:
(Showing results 1 to 4 of 4)


Short Term Energy Forecasting with Neural Networks

J. Stuart McMenamin and Frank A. Monforte

Year: 1998
Volume: Volume19
Number: Number 4
DOI: 10.5547/ISSN0195-6574-EJ-Vol19-No4-2
View Abstract

Abstract:
Artificial neural networks are beginning to be used by electric utilities, to forecast hourly system loads on a day ahead basis. This paper discusses the neural network specification in terms of conventional econometric language, providing parallel concepts for terms such as training, learning, and nodes in the, hidden layer. It is shown that these models are flexible nonlinear equations that can be estimated using nonlinear least squares. It is argued that these models are especially well suited to hourly load forecasting, reflecting the presence of important nonlinearities and variable interactions. The paper proceeds to show how conventional statistics, such as the BIC and MAPE statistics can be used to select the number of nodes in the hidden layer. It is concluded that these models provide a powerful, robust and sensible approach to hourly load forecasting that will provide modest improvements in forecast accuracy relative to well-specified regression models.

Purchase ( $25 )

Forecasting Nonlinear Crude Oil Futures Prices

Saeed Moshiri and Faezeh Foroutan

Year: 2006
Volume: Volume 27
Number: Number 4
DOI: 10.5547/ISSN0195-6574-EJ-Vol27-No4-4
View Abstract

Abstract:
The movements in oil prices are very complex and, therefore, seem to be unpredictable. However, one of the main challenges facing econometric models is to forecast such seemingly unpredictable economic series. Traditional linear structural models have not been promising when used for oil price forecasting. Although linear and nonlinear time series models have performed much better in forecasting oil prices, there is still room for improvement. If the data generating process is nonlinear, applying linear models could result in large forecast errors. Model specification in nonlinear modeling, however, can be very case dependent and time-consuming.In this paper, we model and forecast daily crude oil futures prices from 1983 to 2003, listed in NYMEX, applying ARIMA and GARCH models. We then test for chaos using embedding dimension, BDS(L), Lyapunov exponent, and neural networks tests. Finally, we set up a nonlinear and flexible ANN model to forecast the series. Since the test results indicate that crude oil futures prices follow a complex nonlinear dynamic process, we expect that the ANN model will improve forecasting accuracy. A comparison of the results of the forecasts among different models confirms that this is indeed the case.

Purchase ( $25 )

Size Matters: Estimation Sample Length and Electricity Price Forecasting Accuracy

Carlo Fezzi and Luca Mosetti

Year: 2020
Volume: Volume 41
Number: Number 4
DOI: 10.5547/01956574.41.4.cfez
View Abstract

Abstract:
Short-term electricity price forecasting models are typically estimated via rolling windows, i.e. by using only the most recent observations. Nonetheless, the literature does not provide guidelines on how to select the optimal size of such windows. This paper shows that determining the appropriate window prior to estimation dramatically improves forecasting performances. In addition, it proposes a simple two-step approach to choose the best performing models and window sizes. The value of this methodology is illustrated by analyzing hourly datasets from two large power markets (Nord Pool and IPEX) with a selection of eleven different forecasting models. Incidentally, our empirical application reveals that simple models, such as a simple linear regression (SLR) with only two parameters, can perform unexpectedly well if estimated on extremely short samples. Surprisingly, in the Nord Pool, such SLR is the best performing model in 13 out 24 trading periods.

Purchase ( $25 )

Drilling Deeper: Non-Linear, Non-Parametric Natural Gas Price and Volatility Forecasting

Dusan Bajatovic, Deniz Erdemlioglu, and Nikola Gradojevic

Year: 2024
Volume: Volume 45
Number: Number 4
DOI: 10.5547/01956574.45.4.dbaj
View Abstract

Abstract:
This paper studies the forecast accuracy and explainability of a battery of day-ahead (Henry Hub and Title Transfer Facility (TTF)) natural gas price and volatility models. The results demonstrate the dominance of non-linear, non-parametric models with deep structure relative to various competing model specifications. By employing the explainable artificial intelligence (XAI) approach, we document that the price of natural gas is formed strategically based on crude oil and electricity prices. While the conditional volatility of natural gas returns is driven by long-memory dynamics and crude oil volatility, the informativeness of the electricity predictor has improved over the most recent volatile time period. Although we reveal that predictive non-linear relationships are inherently complex and time-varying, our findings in general support the notion that natural gas, crude oil and electricity are interconnected. Focusing on the periods when markets experienced sharp structural breaks and extreme volatility (e.g., the COVID-19 pandemic and the Russia-Ukraine conflict), we show that deep learning models provide better adaptability and lead to significantly more accurate forecast performance.

Purchase ( $25 )



Begin New Search
Proceed to Checkout

 

© 2024 International Association for Energy Economics | Privacy Policy | Return Policy