Scholarlink

<p>The complex dynamics of global development have turned crude oil from a critical strategic commodity into a powerful tool for political‒military influence. This study aims to forecast the price of OPEC crude oil by simultaneously considering the complex interaction between financial markets, the multifaceted political landscape of the Middle East, and the inherent complexity of oil markets. What they all have in common is their inherent uncertainty. To address this challenge, since this work seeks to identify a robust set of explanatory variables affecting oil prices, the adaptive neural-fuzzy inference system (ANFIS) was employed. A set of input variables with different points of view, which are the time series of monthly data from 2001 to 2019, are considered. In ANFIS, the Sugeno system and the C-Means clustering method are used. The results show the influence of different oil markets on each other, and among the unconventional variables, economic policy uncertainty is not influential, whereas terrorist attacks and geopolitical risk influence it. The numerical results following the implementation of the proposed model show a significant increase of 0.4417 compared with the conventional modeling approaches in this field. This discrepancy highlights the advantages of the proposed methodology.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Investigating the impact of unconventional variables on the improvement of OPEC crude oil price forecasting modeling

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年2月3日

查看全文
  • S. A. Edalatpanah,
  • S. Fatemeh Faghidian,
  • Dragan Pamucar,
  • Vladimir Simic

摘要

The complex dynamics of global development have turned crude oil from a critical strategic commodity into a powerful tool for political‒military influence. This study aims to forecast the price of OPEC crude oil by simultaneously considering the complex interaction between financial markets, the multifaceted political landscape of the Middle East, and the inherent complexity of oil markets. What they all have in common is their inherent uncertainty. To address this challenge, since this work seeks to identify a robust set of explanatory variables affecting oil prices, the adaptive neural-fuzzy inference system (ANFIS) was employed. A set of input variables with different points of view, which are the time series of monthly data from 2001 to 2019, are considered. In ANFIS, the Sugeno system and the C-Means clustering method are used. The results show the influence of different oil markets on each other, and among the unconventional variables, economic policy uncertainty is not influential, whereas terrorist attacks and geopolitical risk influence it. The numerical results following the implementation of the proposed model show a significant increase of 0.4417 compared with the conventional modeling approaches in this field. This discrepancy highlights the advantages of the proposed methodology.