Intuitionistic fuzzy time series forecasting model based on intuitionistic fuzzification function approach

• Published in: AIP Conference Proceedings
• Main Author: Alam N.M.F.H.N.B.; Ramli N.; Nassir A.A.; Mohd A.H.; Mohammed N.
• Format: Conference paper
• Language: English
• Published: American Institute of Physics Inc. 2024
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85182563548&doi=10.1063%2f5.0171650&partnerID=40&md5=be1d8e664ddf5accf6f0aa1c81999c79

Fuzzy time series is a well-known forecasting method that can deal with data in linguistic forms. The fuzzy time series was extended into intuitionistic fuzzy sets (IFS), which can handle non-determinacy in time series forecasting more efficiently. Many forecasting models based on the IFS have been proposed using Atanassov's conversion methods; however, the hesitation index was ignored during the establishment of IFS. In this regard, this study develops an intuitionistic fuzzy time series forecasting model using the intuitionistic fuzzification functions, which consider the hesitation index in the establishment of IFS. The developed model was implemented in predicting Malaysian crude palm oil prices from January 2017 until December 2021. The result shows that the proposed model exhibits a better forecasting performance compared to the existing models based on Atanassov's conversion method. In the proposed model, the nature of IFS is preserved since the membership, non-membership, and hesitation degrees are involved in the forecasting process. Based on the strength of the intuitionistic fuzzification functions, the model can be improved further by modifying the intuitionistic defuzzification method in the future. © 2024 Author(s).