Development of a Weighted Fractional Time-Varying Discrete Grey Model for CO2 Emissions Forecasting

Ani Shabri
dx.doi.org/10.17576/apjitm-2026-1501-14

Abstract

The data accumulation operators are the basic elements of the grey system theory and they help in transforming the raw data sequences into other forms whereby more clear patterns might be understood. Nonetheless, each of the accumulation methods is usually applicable to particular forms of time series activity. In order to exploit the advantages of different accumulation techniques even better, especially those which are in a position to detect nonlinear patterns, this paper presents a new model named the Time-Varying Discrete Grey Model (TDGM) with Weighted Fractional Accumulation. The best parameter values are determined effectively and prediction errors are reduced with the help of the Quantum Particle Swarm Optimization (QPSO) algorithm. To test the effectiveness of the model, four case studies of CO2 emission in Indonesia, Malaysia, Singapore, and Thailand were used. These findings are clear to hasten that the new model is capable of producing better prediction and produce lower error rates of previous models and able to identify complex trend nonlinearity in a real environmental data.

keyword

CO2 emission, Grey model, ARIMA, SVM, forecasting

Area

Data Mining and Optimization