The impact of Hurricanes Gustav and Ike on offshore oil and gas production in the Gulf of Mexico

“…The international crude oil prices are influenced by many external factors in addition to historical price fluctuations, making the accurate and reliable forecasting a difficult task. The empirical results in Demirer and Kutan (2010) and Kaiser and Yu (2010) suggest that incorporating UCG information from the social media into crude oil price forecasting can achieve better performance. To obtain higher accuracy, researchers add more various exogenous variables to their forecasting framework, such as Li et al (2019).…”

Section: Discussionmentioning

confidence: 98%

“…Methods for processing text data have constantly appeared (e.g., Aggarwal & Zhai, 2012;Berry & Castellanos, 2004;Shriharir & Desai, 2015) and got increasingly mature. Some studies have suggested that the information extracted from the UGC can contribute to the prediction of financial data (Demirer & Kutan, 2010;Kaiser & Yu, 2010). Online news is an important part of UGC and contains rich and valuable information which can be utilized to quantify the changes of the public's mood (Serrano-Guerrero, Olivas, Romero, & Herrera-Viedma, 2015) and the market.…”

Section: Introductionmentioning

confidence: 99%

Crude oil price forecasting incorporating news text

Bai

Hao

et al. 2022

International Journal of Forecasting

View full text Add to dashboard Cite

Sparse and short news headlines can be arbitrary, noisy, and ambiguous, making it difficult for classic topic model LDA (latent Dirichlet allocation) designed for accommodating long text to discover knowledge from them. Nonetheless, some of the existing research about text-based crude oil forecasting employs LDA to explore topics from news headlines, resulting in a mismatch between the short text and the topic model and further affecting the forecasting performance. Exploiting advanced and appropriate methods to construct high-quality features from news headlines becomes crucial in crude oil forecasting. This paper introduces two novel indicators of topic and sentiment for the short and sparse text data to tackle this issue. Empirical experiments show that AdaBoost.RT with our proposed text indicators, with a more comprehensive view and characterization of the short and sparse text data, outperforms the other benchmarks. Another significant merit is that our method also yields good forecasting performance when applied to other futures commodities.

show abstract

“…Numerous studies [5]- [7] argued that additional market factors should be added into the analytical framework for crude oil prices. It is indicated that oil prices have been sensitive to the oil-related events, such as invasion, extreme weather and OPEC production quota decisions etc.…”

Section: Introductionmentioning

confidence: 99%

BEMD-Based Event Analysis for Crude Oil Market: Evidence Using Google Search Data

Wang

Chen

2015

2015 IEEE International Conference on Smart City/SocialCom/SustainCom (SmartCity)

View full text Add to dashboard Cite

The influences of extreme events on crude oil market is of great significance in crude oil price analysis and the growing Internet concern about emergencies often accompanies panic of the market traders, which would probably result in turmoil and instability of the oil market. In this paper, a novel modeling framework that integrates Bivariate Empirical Mode Decomposition (BEMD) and Autoregressive-Generalized Autoregressive Conditional Heteroskedasticity (AR-GARCH) is proposed for analyzing the effects of three types of oil-related events (abnormal climate, economic crisis, war) on crude oil market. The Internet concern derived by search volumes from Google Trends is introduced to characterize the magnitude and dynamic path of the response to oil-related events. In our approach, the original time series is decomposed into several intrinsic modes with different time scale, capturing the fluctuations caused by extreme events. The empirical results indicate that oil prices responding to different types of oil-related events present differentiation. Meanwhile, the effects of various intrinsic modes are also different, which implies distinguishing conduction ways and effects with different frequencies. By incorporating open source data into BEMD-based event analysis, this paper provides a feasible solution to estimate the impacts of extreme events on crude oil market.

show abstract

The impact of Hurricanes Gustav and Ike on offshore oil and gas production in the Gulf of Mexico

Cited by 20 publications

References 4 publications

Oil price volatility and oil-related events: An Internet concern study perspective

Oil price volatility and oil-related events: An Internet concern study perspective

Crude oil price forecasting incorporating news text

BEMD-Based Event Analysis for Crude Oil Market: Evidence Using Google Search Data

Contact Info

Product

Resources

About