Forecasting Applied to the Electricity, Energy, Gas and Oil Industries: A Systematic Review

Abstract

This paper presents a literature review in which methodologies employed for the forecast of the price of stock companies and raw materials in the fields of electricity, oil, gas and energy are studied. This research also makes an analysis of which data variables are employed for training the forecasting models. Three scientific databases were consulted to perform the present research: The Directory of Open Access Journals, the Multidisciplinary Digital Publishing Institute and the Springer Link. After running the same query in the three databases and considering the period from January 2017 to December 2021, a total of 1683 articles were included in the analysis. Of these, only 13 were considered relevant for the topic under study. The results obtained showed that when compared with other areas, few papers focus on the forecasting of the prices of raw materials and stocks of companies in the field under study. Furthermore, most make use of either machine learning methodologies or time series analysis. Finally, it is also remarkable that some not only make use of existing algorithms but also develop and test new methodologies.

Stock Price Forecasting of IBEX35 Companies in the Petroleum, Electricity, and Gas Industries

“…Despite this concept and taking into account that this hypothesis would be false, many authors do not believe that markets are entirely efficient [18,19], especially considering phenomena such as their tendency to overreact in the short term [20,21]. Additionally, several of the key premises of the EMH have been shown to be false [22]. Considering these facts and the large amount of money present in the stock markets, the growing interest in the study of market dynamics and the creation of forecasting models makes sense.…”

“…In recent years, time series forecasting has become an essential tool for stock market analysts to make informed decisions regarding stock prices [22]. One of the significant benefits of using time series forecasting in the stock market is its ability to predict future stock prices.…”

“…Another benefit of time series forecasting is its ability to identify long-term trends in the stock market [24]. By analyzing historical data over a more extended period, analysts can identify long-term trends in the market and use this information to develop long-term investment strategies [22].…”

“…One of the most significant of these is that it relies on historical data, which may not be an accurate representation of future market behavior. The stock market is subject to many external factors, such as economic conditions, political events, and company-specific news, which can affect stock prices in unpredictable ways [22]. This means that historical data may not always be a reliable predictor of future trends.…”

“…The purpose of this research is not only to develop forecasting models but also to analyze stock price dynamics under different market conditions. Due to the complexity of financial decisions in a changing environment, individuals who make them must consider several variables that require sophisticated mathematical models for their analysis [22]. Therefore, the fields of multivariate data analysis and machine learning have produced some of the most promising tools used in this undertaking.…”