Simulation of Dynamic Prediction Model for the Growth of Live E-commerce Trading Scale Based on Grey Clustering Algorithm

In today’s increasingly globalized world, traditional trade needs to transform its past business model and leverage the correlation between live streaming EC (Electronic Commerce) and trade growth to achieve interaction between the two. This is an urgent issue that EC needs to address. This article attempts to conduct research on the dynamic prediction model simulation of live streaming EC transaction size growth based on grey clustering algorithm. The grey clustering algorithm quantitatively analyzes the balance and stability of the model, improves optimization and releases scientific and reasonable plans, which is of great significance for the healthy and stable development of the EC market. The research results show that compared with ACA (Ant Colony Algorithm) and PSO (Particle Swarm optimization) algorithms and the algorithm in this paper, the Root-mean-square deviation of the algorithm in this paper is 0.141, and the relative error is 0.119. Both have better error values than ACA and PSO. It can be seen that the live EC transaction volume estimation of the algorithm in this article can update and estimate all components of the transaction volume state variable, effectively ensuring the quality of the algorithm’s live EC transaction volume estimation.