Forecasting China's SO2 emissions by the nonlinear grey Bernoulli self-memory model (NGBSM)

Abstract

The paper presents a novel nonlinear grey Bernoulli self-memory model (NGBSM) for the data sequences characteristics of saturation or fluctuation. The NGBSM model combines the advantages of the self-memory principle of dynamic systems and the traditional nonlinear grey Bernoulli model through a coupling of the above two prediction methods. The weakness of the traditional grey prediction model, i.e., being sensitive to the initial value, can be overcome by using a multi-time-point initial field instead of only a single-time-point initial field in the system's self-memorization equation. As shown in the case study of China's SO2 emissions, the NGBSM model can take full advantage of the system's multi-time historical data and accurately predict the system's evolutionary trend. Three popular accuracy check criteria are adopted to test and verify the reliability and stability of the NGBSM model, and its superior predictive performance over other traditional grey prediction models. The results show that the proposed NGBSM model enriches grey prediction theory, and can be applied to other similar data sequences.