Chemical commodity price forecast based on multi-factor combination model

Abstract

Chemical products are the cornerstone of agricultural development and provide important raw materials for industrial and agricultural production. Their quality, quantity and price stability are closely watched by upstream and downstream industries. Therefore, it is of great practical and theoretical significance to accurately predict the prices of various chemical products. Due to the influence of various factors on the price of chemical products, they show strong nonlinear, non-stationary and no obvious trend characteristics. It is difficult for the traditional single model to capture the internal hidden laws of their data. Based on this, this paper proposes a combination forecasting model based on data decomposition, which can deeply mine the potential volatility characteristics within the data, so as to better grasp the volatility law and realize its price forecast. Guided by the idea of "decomposing input and combining output," this paper constructs a price forecasting model based on multi factor decomposition and integration, and explains and forecasts it from the factor level.