A New Forecasting Approach for Oil Price Using the Recursive Decomposition-Reconstruction-Ensemble Method with...

Area of Science:

• Economics
• Financial Markets
• Time Series Analysis

Background:

• Crude oil price forecasting is critical due to its broad economic impact.
• Existing forecasting methods often struggle with the complex, non-linear nature of oil price data.

Purpose of the Study:

• To propose a novel recursive decomposition-reconstruction-ensemble model for enhanced crude oil price forecasting.
• To address the complexity and dynamic traits inherent in oil price data.

Main Methods:

• Utilized Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) for data decomposition.
• Developed a recursive model incorporating mode reconstruction, component prediction, and ensemble prediction driven by complexity traits.
• Validated the model using West Texas Intermediate (WTI) and Brent crude oil spot prices.

Main Results:

• The proposed recursive CEEMDAN decomposition-reconstruction-ensemble model exhibited superior prediction performance.
• Empirical results confirmed the model's effectiveness over benchmark forecasting techniques.
• The model successfully captured the complexity traits of crude oil price data.

Conclusions:

• The recursive CEEMDAN decomposition-reconstruction-ensemble model is an effective tool for future oil price forecasting.
• This approach offers improved accuracy and reliability for predicting volatile crude oil markets.
• The findings support the adoption of this advanced method by academics and policymakers.