A Neutrosophic Forecasting Model for Time Series Based on First-Order State and Information Entropy of High-Order...

Area of Science:

• Data Science
• Financial Forecasting
• Time Series Analysis

Background:

• Traditional time series models overlook fluctuation inconsistency.
• Existing methods lack a comprehensive representation of historical data dynamics.

Purpose of the Study:

• To propose a novel time series forecasting model that incorporates fluctuation inconsistency.
• To enhance prediction efficiency by quantifying and utilizing data fluctuation characteristics.

Main Methods:

• Converting time series to fluctuation time series.
• Mapping trends to neutrosophic set memberships (truth, falsity, indeterminacy).
• Utilizing information entropy to quantify fluctuation inconsistency.
• Applying neutrosophic similarity and weighted arithmetic averaging for forecasting.

Main Results:

• The proposed neutrosophic forecasting model based on information entropy (NFM-IE) effectively represents both fluctuation trends and consistency.
• Experimental results on TAIEX, SHSECI, and HSI datasets show stable and accurate predictions.
• The NFM-IE model outperforms existing approaches in prediction accuracy and universality.

Conclusions:

• The NFM-IE model offers a significant advancement in time series forecasting by integrating fluctuation inconsistency.
• The model's ability to handle diverse datasets highlights its practical applicability.
• This approach provides a robust framework for future time series prediction research.