Oil Price Forecasting Research Articles

Analysis and forecasting of crude oil price based on the variable selection-LSTM integrated model

In recent years, the crude oil market has entered a new period of development and the core influence factors of crude oil have also been a change. Thus, we develop a new research framework for core influence factors selection and forecasting. Firstly, this paper assesses and selects core influence factors with the elastic-net regularized generalized linear Model (GLMNET), spike-slab lasso method, and Bayesian model average (BMA). Secondly, the new machine learning method long short-term Memory Network (LSTM) is developed for crude oil price forecasting. Then six different forecasting techniques, random walk (RW), autoregressive integrated moving average models (ARMA), elman neural Networks (ENN), ELM Neural Networks (EL), walvet neural networks (WNN) and generalized regression neural network Models (GRNN) were used to forecast the price. Finally, we compare and analyze the different results with root mean squared error (RMSE), mean absolute percentage error (MAPE), directional symmetry (DS). Our empirical results show that the variable selection-LSTM method outperforms the benchmark methods in both level and directional forecasting accuracy.

How good are analyst forecasts of oil prices?

Even though there is a wide consensus that having good oil price forecasts is very valuable for many agents in the economy, results have not been fully satisfactory and there is an ongoing effort to improve their accuracy. Research has explored many different modeling approaches including time series, regressions, and artificial intelligence, among others. Also, many different sources of input data have been used like spot and futures prices, product spreads, and micro and macro variables.This paper explores how useful analyst expected price data are for forecasting when appropriate measures are taken to account for their sparse nature and high volatility. It proposes a multifactor stochastic pricing model, with time-varying risk premiums calibrated with filtered futures and analyst forecasts using a Kalman Filter.The forecasting model is applied to ten years of oil prices and analyst forecasts, from NYMEX and Bloomberg, respectively. Results are very encouraging showing that the model forecasts are much better than the no-change forecasts, commonly used as a benchmark, and better than those from the widely used Bloomberg's Consensus Expected Price Model. We conclude that analyst forecasts are a valuable source of input data that should be considered in future forecasting models.

The Financialization of Crude Oil Markets and Its Impact on Market Efficiency: Evidence from the Predictive Ability and Performance of Technical Trading Strategies

Oil price forecasts are of crucial importance for many policy institutions, including the European Central Bank and the Federal Reserve Board, but projecting oil market evolutions remains a complicated task, further exacerbated by the financialization process that characterizes the crude oil markets. The efficiency (in Fama’s sense) of crude oil markets is revisited in this research through the investigation of the predictive ability of technical trading rules (TTRs). The predictive ability and trading performance of a plethora of TTRs are explored on the crude oil markets, as well as on the energy sector ETF XLE, while taking a special focus on the turbulent COVID-19 pandemic period. We are interested in whether technical trading strategies, by signaling the right timing of market entry and exits, can predict oil market movements. Research findings help to confidently conclude on the weak-form efficiency of the WTI crude oil and the XLE fund markets throughout the 1999–2021 period relative to the universe of TTRs. Moreover, results attest that TTRs do not add value to the Brent market beyond what may be expected by chance over the pre-pandemic 1999–2019 period, confirming the efficiency of the market before 2020. Nonetheless, research findings also suggest some temporal inefficiency of the Brent market during the 1 and ¼ years of pandemic period, with important consequences for energy markets’ practitioners and issuers of policy. Research findings further imply that there is evidence of a more intense financialization of the WTI crude oil market, which requires tighter measures from regulators during distressed markets. The Brent oil market is affected mainly by variations in oil demand and supply at the world level and to a lesser degree by financialization and the activity of market practitioners. As such, we conclude that different policies are needed for the two oil markets and also that policy issuers should employ distinct techniques for oil price forecasting.

A memory-trait-driven decomposition–reconstruction–ensemble​ learning paradigm for oil price forecasting

In order to improve the prediction performance in oil price forecasting, a novel memory-trait-driven decomposition–reconstruction–ensemble learning paradigm is proposed for oil price forecasting. The proposed methodology consists of four steps, i.e., data decomposition for original complex time series, component reconstruction for decomposed components, individual prediction for the reconstructed components, and ensemble output based on the individual component prediction results, which are all driven by memory traits. For verification purpose, the West Texas Intermediate (WTI) crude oil spot prices are used as the sample data. The experimental results demonstrated that the proposed methodology can produce the better and more robust results relative to the benchmarking models listed in this study. This indicates that the proposed memory-trait-driven decomposition–reconstruction–ensemble​ methodology can be used as a promising solution to oil price prediction with the traits of memory.

Forecasting crude oil price intervals and return volatility via autoregressive conditional interval models

Crude oil prices are of vital importance for market participants and governments to make energy policies and decisions. In this paper, we apply a newly proposed autoregressive conditional interval (ACI) model to forecast crude oil prices. Compared with the existing point-based forecasting models, the interval-based ACI model can capture the dynamics of oil prices in both level and range of variation in a unified framework. Rich information contained in interval-valued observations can be simultaneously utilized, thus enhancing parameter estimation efficiency and model forecasting accuracy. In forecasting the monthly West Texas Intermediate (WTI) crude oil prices, we document that the ACI models outperform the popular point-based time series models. In particular, ACI models deliver better forecasts than univariate ARMA models and the vector error correction model (VECM). The gain of ACI models is found in out-of-sample monthly price interval forecasts as well as forecasts for point-valued highs, lows, and ranges. Compared with GARCH and conditional autoregressive range (CARR) models, ACI models are also superior in volatility (conditional variance) forecasts of oil prices. A trading strategy that makes use of the monthly high and low forecasts is further developed. This trading strategy generally yields more profitable trading returns under the ACI models than the point-based VECM.

Risk measurement of oil price based on Bayesian nonlinear quantile regression model

Oil price forecasting is one of the most challenging issues since it is noisy, non-stationary, and chaotic. In this paper, we design a Bayesian Nonlinear Quantile method consisting of a Threshold Improved model and an Adaptive MCMC model to improve predicting performance. Specifically, the threshold improve model is introduced to solve the problems caused by the completely asymmetric distribution, and the Adaptive MCMC model is used to get the optimal threshold. Besides, the two-stage framework is applied to improve traditional methods' performance, including the Indirect GARCH model and Asymmetric Slope model. The experimental results show that our approach provides a promising alternative to oil price prediction, and the framework also improve the performance of the traditional methods. The contribution of this paper is to improve the accuracy of the oil price forecasting model, and the framework applies to other energy prices as well.

Forecasting the price of crude oil

Crude oil is the mixture of petroleum liquids and gases that is extracted from the ground by oil wells. It is an important source of fuel and is used in the production of several products. Given the important role price of the crude oil plays, it becomes extremely important for managers to predict future oil price while making operational decisions such as: when to purchase material, how much to produce and what modes of transportation to use. The goal of this paper is to develop a forecasting model to predict the oil prices that aid management to reduce operational costs, increase profit and enhance competitive advantage. We first analyze the primary theories related to the forecast of oil price followed by the reviews of two main streams of forecast theory, which are Target Capacity Utilization Rule (TCU) and Exhaustible Resources Theory. We implement a Target Capacity Utilization Rule recursive simulation model and test it on the historical data from 1987 through 2017 to predict crude oil prices for 1991 through 2017. We tried several variations of the base model and the best method produced MAD, MSE, MAPE and MPE of 12.676, 280.92, 0.2597, 0.028, respectively. We further estimated the forecasts of the oil prices at a monthly level based on our yearly forecast of oil prices from our best method. The calculated MAD, MSE, MAPE and MPE values are 5.66, 82.1163, 0.1246 and 0.038, respectively, which shows our model is promising again at a monthly level.

A New Approach for Forecasting Crude Oil Prices Using Median Ensemble Empirical Mode Decomposition and Group Method of Data Handling

The accuracy of time series forecasting is more important and can assist organizations to take up-to-date decisions for better planning and management. Several classical econometrics and computational approaches show promising results for the ordinary time series forecasting tasks, but they are not satisfactory in crude oil price forecasting. Ensemble empirical mode decomposition (EEMD) not only resolves the problem of nonlinearity and nonstationarity of time series prediction but also creates some problems (i.e., mood mixing and splitting). In this study, we proposed a new hybrid method that combines the median ensemble empirical mode decomposition and group method of data handling (MEEMD-GMDH) to reduce mood splitting problems and forecast crude oil price. MEEMD is achieved by replacing the mean operator with the median operator during the EEMD process. For testing and validation purposes of the different models, the two-seat stamp benchmarked crude oil price data are used (i.e., Brent and West Texas Intermediate (WTI)). To check the proposed model performance, different evaluation measures are used including Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and Diebold-Mariano (DM) test. All the forecasting accuracy measures confirmed that our proposed model performs well in crude oil prices forecasting as compared to other hybrid models.

A combined architecture of multivariate LSTM with Mahalanobis and Z-Score transformations for oil price forecasting

Oil price plays a vital role in a country's economy. Oil price forecasting helps in making better economic planning and decisions. The fluctuation in the oil price occurs due to several factors that make oil price forecasting a challenging task. Deep learning models such as Recurrent Neural Networks (RNN) have emerged as a successful method in solving many nonlinear and complex problems. In this research, a combined architecture of Multivariate Long Short Term Memory (MLSTM) is proposed with Mahalanobis and Z-score transformations. These transformations improve the data to uncorrelated and standardized variance, thus making data more suitable for regression analysis. The available historical time-series data on the West Texas Intermediate (WTI) oil prices and the factors affecting the oil prices are collected to form the data set. The feature selection process using selectKBest and correlation analysis is carried out. We develop six variants of MLSTM models with different features, transformations, and outlier elimination. The MLSTM model with Z-score transformation on the selected features and outlier eliminated data gives an RMSE of 0.212 and an R2 score of 0.954. This indicates that our model performed well.

Does News Tone help forecast Oil?

Does news tone help forecast oil? In this paper, we study the relationship between news tone and crude oil prices and evaluate the role news tone plays in the ability to forecast oil prices. Specifically, we use the oil-specific dictionary and the approach proposed in the study of Loughran, as well as a competing general financial dictionary of Henry, to directly measure the sentiment of 3579 oil news articles from Financial Times for actual oil price forecasting. We find compelling evidence that news tone constructed by the oil dictionary helps forecast monthly oil prices out-of-sample over short horizons, while the news tone constructed by the financial dictionary shows no out-of-sample forecasting power at all. We verify and document the economic significance of the best performing forecasting model against the others and a naive buy-hold strategy. We argue that the forecasting power of news tone is data and method dependent, and we underscore the correct use of domain-specific dictionaries in financial sentiment analysis.

Bayesian Structural Time Series for Forecasting Oil Prices

There are many methods of forecasting, and these methods take data only, analyze it, make a prediction by analyzing, neglect the prior information side and do not considering the fluctuations that occur overtime. The best way to forecast oil prices that takes the fluctuations that occur overtime and is updated by entering prior information is the Bayesian structural time series (BSTS) method. Oil prices fluctuations have an important role in economic so predictions of future oil prices that are crucial for many countries whose economies depend mainly on oil, such as Iraq. Oil prices directly affect the health of the economy. Thus, it is necessary to forecast future oil price with models adapted for emerging events. In this article, we study the Bayesian structural time series (BSTS) for forecasting oil prices. Results show that the price of oil will increase to 156.2$ by 2035.

A MODEL SELECTION FOR PRICE FORCASTING OF CRUDE PALM OIL AND FRESH FRUIT BUNCH PRICE FORECASTING

This study was aimed to determining a fitted forecasting method for the forecasting of crude palm oil prices at international and domestic market as well as fresh fruit bunch prices at collecting merchant and farmer level in Bengkulu Province market by considering three models, namely, double exponential smoothing, autoregressive integrated moving average, and classical decomposition. The data used were monthly data of crude palm oil prices at domestic and world markets from January 2012 – October 2016 and January 2012 – April 2017, while the fresh fruit bunch data at collecting merchant and farmers in Bengkulu Province were also monthly data from 2007 – 2014. The result showedthat the most accurate method was ARIMA for all prices at all market levels. This decision was based on all criteria used to determine the best model including MAPE, MAD, and MSD.

Prospects for using data from pending orders for forecasting oil prices in Russia

Modern oil-pricing is more dependent on the stock market conditions than on the cost price or demand and supply laws. The price is infl uenced by a great number of objective and subjective factors. The ability to analyze these factors is the basis of the modern stock market trade. Thus, stock quotes are the key to analyzing oil market in the short-term perspective. The authors study current trends informing pricing factors of the oil market and their influence. They point out the peculiarities of using orders as pricing tools, adduce Russian experience on pending orders at the oil market, analyze the specificity and prospects of using pending orders for analyzing oil market. So, changes of the conditions can be predicted much faster than by using traditional statistic methods of analyzing transactions. The authors conclude that using such a tool as a pending order will lead to better understanding of the market conditions for the main Russian oil brand (Urals). It will make the market more predictable and controllable by the government which will mitigate the consequences of drastic changes in oil prices and the changes of the cost of energy and fuel arising from that making the economy more resistant to crises. More accurate forecasts will increase the Russian traders’ income from the transactions. The authors use international researches devoted to stock market trade and data analysis, and information from software developers who design programs for analyzing stock market data.

Coping with a dual shock: The economic effects of COVID-19 and oil price crises on African economies

Oil-dependent countries face a twin-shock: in addition to the COVID-19 outbreak, they are facing an oil price collapse. In this paper, we study the impact of this dual shock on the forecasted GDP growth in Africa using the COVID-19 outbreak as a natural experiment. We use the IMF World Economic Outlook’s GDP growth forecasts before and after the outbreak. We find that COVID-19 related deaths result in -2.75 percentage points forecasted GDP growth loss in the all sample while oil-dependence induces -7.6 percentage points loss. We document that the joint shock entails higher forecasted growth loss in oil-dependent economies (-10.75 percentage points). Based on oil price forecasts and our empirical findings, we identify five recovery policies with high potential: social safety net policy, economic diversification, innovation and technological transformation, fiscal discipline, and climate-friendly recovery policy.

Forecasting the Price of Fuel Oil: A STL-(ELM+ARIMA) Combination Approach

Focusing on fuel oil price forecasting, we propose a “decomposition-prediction-integration” route and STL-(ELM+ARIMA) combination forecasting model. This model decomposes the fuel oil price time series by STL, and effectively combines the advantages of high frequency seasonal cycle and short-term fluctuation time series forecasting in ELM non-parametric model with the advantages of low frequency trend forecasting in ARIMA parametric model. Finally, this paper conducts an empirical study on the spot price of Singapore’s Platts fuel oil 180CST to verify the effectiveness of the proposed forecasting method. The results show that the forecasting accuracy of 180 CST fuel oil price model based on STL-( ELM (1) + ARIMA (2) +ELM (3)) is highest.

Forecasting the U.S. oil markets based on social media information during the COVID-19 pandemic

Accurate oil market forecasting plays an important role in the theory and application of oil supply chain management for profit maximization and risk minimization. However, the coronavirus disease 2019 (COVID-19) has compelled governments worldwide to impose restrictions, consequently forcing the closure of most social and economic activities. The latter leads to the volatility of the oil markets and poses a huge challenge to oil market forecasting. Fortunately, the social media information can finely reflect oil market factors and exogenous factors, such as conflicts and political instability. Accordingly, this study collected vast online oil news and used convolutional neural network to extract relevant information automatically. Oil markets are divided into four categories: oil price, oil production, oil consumption, and oil inventory. A total of 16,794; 9,139; 8,314; and 8,548 news headlines were collected in four respective cases. Experimental results indicate that social media information contributes to the forecasting of oil price, oil production and oil consumption. The mean absolute percentage errors are respectively 0.0717, 0.0144 and 0.0168 for the oil price, production, and consumption prediction during the COVID-19 pandemic. Marketers must consider the impact of social media information on the oil or similar markets, especially during the COVID-19 outbreak.

Overview of the ARIMA Model Average Crude Oil Price Forecast and its Im-plications on the Indian Economy Post-Liberalization

The crude oil markets have seen unpredictability over petroleum markets' good and bad times worldwide in the past few years. Various facets of the natural energy industry discuss in this research paper. The effect of the Indian ecosystem in determining oil demand and supply is vital to understand price fluctuations. Researchers in this research paper concentrates on the ARIMA model and other regression models used in the post-1991 LPG reforms to determine crude petroleum values and their primary effect on the Indian ecosystem (GDP) through time-series data from 1991 to 2019. The ARIMA model is further evidence of the validation of datasets and the potential trend in the show of global oil rates. The uncouth oil historical values prophesy the potential movement from crude oil prices, and the recommended model-based fall-outs are compelling and honest. Before applying the normality of data and unit-roots, the ARIMA model thoroughly checks at the level. If stationary is not detected, then unit-roots are suspended. The first step is to keep residuals static at the required level of significance to provide efficient forecasting using the ARIMA model used for this analysis.

MULTI-SOURCE MODEL OF HETEROGENEOUS DATA ANALYSIS FOR OIL PRICE FORECASTING

This article sheds light on the question of whether it is possible to create fairly accurate forecasts of real oil prices. For this purpose, a multi-level machine learning model has been created to analyze several sources of heterogeneous data to predict future prices. The article uses different types of data: market condition data, titles, and transaction data. Then, they have been processed to be able to load them into the model. The validation of the regression neural network results showed that the model is more accurate than in previous studies. In fact, this paper presents an artificial neural network model that solves the problem of determining the most informative relationship between different types of oil price data. Keywords: artificial neural network, oil forecasting, machine learning, price prediction, energy resources.JEL Classifications: C45, C51, Q43, Q47DOI: https://doi.org/10.32479/ijeep.10853

An Empirical Investigation into the Effect of Global Oil Price on Nigeria Gross Domestic Product from (2000-2019)

Nigeria is one of the largest oil producing countries in the world, its revenue is largely dependent on crude oil sales. The country vulnerability to crude oil price fluctuations is a phenomenon that has been a reoccurring event Thus; this study aims to examine the impact of Global crude oil prices on Nigeria gross domestic product (GDP) and the economy at large. The study relied deeply on secondary data obtained from the Central bank of Nigeria (CBN). The data were subjected to simple regression analysis, Pearson Product Moment Correlation to determine the effect and relationship between oil price (Independent Variable) and Nigeria GDP (dependent variables). And time series analysis was also used to examine the trend in the data set and fit a more parsimonious model that would aid forecasting of future crude oil price and GDP. The study found that Global crude oil price have a significant correlation with Nigeria gross domestic product, %58.1 of variation in Nigeria gross domestic product is attributed to fluctuation in Global crude oil price and Nigeria GDP will increase by ₦4.21bn per unit increase in crude oil price. The study thus conclude that global crude oil price has a significant impact on Nigeria GDP and its economy at large. The study among others recommend diversification of the Nigerian economy is most imperative given the economic recession in the country now and the impending take-over of the transportation industry by electric vehicles. Keywords: Oil price, Gross domestic product, Economy, Price shock, Exchange rate, Nigeria. DOI: 10.7176/EJBM/13-4-03 Publication date: February 28 th 2021

Do extreme shocks help forecast oil price volatility? The augmented GARCH‐MIDAS approach

AbstractExtreme shocks (e.g., wars and financial crises) cause violent fluctuations in crude oil volatility. In this paper, we first propose GARCH models in the framework of MIDAS augmented to include the impacts of extreme shocks on oil price volatility. In‐sample results show that extreme shocks can induce the additional volatility of crude oil. Further, the results from out‐of‐sample clearly indicate that the crude oil volatility is best fitted by the EGARCH‐MIDAS‐ES model, which incorporates asymmetric effects in the short‐term component and the significant effect of extreme shocks in the long‐term component. Additionally, robustness tests confirm that the augmented volatility models can produce better prediction results, both statistically and economically, than the conventional GARCH‐MIDAS model. Furthermore, we verify that negative extreme shocks can cause larger volatility, whereas positive extreme shocks of the same magnitude have smaller effects. Our contribution offers fresh insights into energy price volatility forecasting by considering extreme shocks.