A0665
Title: Interpret how external shocks affect industrial chain using graph machine learning
Authors: Bin Liu - Southwestern University of Finance and Economics (China) [presenting]
Abstract: A quantitative analysis of the development of the industry chain is conducted from the perspective of external shocks. Factors that may impact the performance of the industrial chain have been studied in the literature, such as government regulation, monetary policy, etc. The interest lies in how to quantify the impacts of these shocks on the industrial chain's performance. To achieve this goal, the industrial chain is modeled with a graph neural network (GNN) and node regression is conducted on some financial performance metrics. To capture the effects of external shocks, it is proposed to compute the interaction between shocks and industrial chain features with a cross-attention module and then filter the original node features in the graph regression. Experiments on two real datasets demonstrate that (i) there are significant effects of external shocks on the industrial chain, and (ii) model parameters, including regression coefficients and the attention map, can explain how external shocks affect the performance of the industrial chain.
