Design of new energy market indicator system and dynamic risk assessment based on graph neural network: enhancing market monitoring and forecasting capability

This paper introduces the theory of the risk spillover effect of the energy market and the theory of market risk transmission and tries to quantify the risk spillover effect between China’s carbon market and the energy market by using the risk measurement and volatility model. The daily closing prices of Shenzhen and Hubei carbon emissions trading and the daily closing price of the CSI New Energy Index are selected as the representative products of China’s carbon market and new energy market, respectively. The GARCH model is used for the yield series volatility test to obtain the risk spillover value between the carbon market and the new energy market. Unify the new factors for evaluating energy price fluctuation risk and establish a new evaluation index system for energy price fluctuation. Using each objective evaluation index, the graph neural network algorithm is determined to assess the prediction performance of new energy market risks. The empirical results show that the prediction results of graph neural networks are 3.52%, 0.756, 0.598, and 0.891 for MAPE, MAE, and RMSE, respectively, and the prediction validity is 0.891. The difference between MAPE, MAE, and RMSE for lag one period and lag four periods are 2.02, 0.598, and 0.299, respectively, and the difference in prediction validity is 0.177, and the four indexes have large differences. It shows that the risks and impacts of the new energy financial market are closely linked. When any influencing factors related to the new energy financial market change, the new energy financial market has a high possibility of risk and needs to be warned. The new energy market risk assessment prediction based on graph neural networks can meet this need and improve the prediction ability of the energy market.