Prediction of Bank Credit Ratings using Heterogeneous Topological Graph Neural Networks

TL;DR

This paper introduces a novel heterogeneous topological graph neural network (HTGNN) that combines persistent homology and traditional lending networks to improve bank credit rating predictions, addressing data privacy issues.

Contribution

The study proposes a new method integrating persistent homology with GNNs to enhance credit rating predictions using incomplete interbank connection data.

Findings

HTGNN outperforms existing models on real-world datasets.

Persistent homology effectively captures complex bank relationships.

The approach aids regulatory and investment decision-making.

Abstract

Agencies such as Standard & Poor's and Moody's provide bank credit ratings that influence economic stability and decision-making by stakeholders. Accurate and timely predictions support informed decision-making, regulatory actions, and investor protection. However, a complete interbank connection graph is often unavailable due to privacy concerns, complicating the direct application of Graph Neural Networks (GNNs) for rating prediction. our research utilizes persistent homology to construct a network that captures relationships among banks and combines this with a traditional lending network to create a heterogeneous network that integrates information from both sources, leading to improved predictions. Experiments on a global, real-world dataset validate the effectiveness of HTGNN. This research has implications for investors and regulatory bodies in enhancing proactive risk mitigation and the implementation of effective market interventions.The code can be find at https://github.com/Liu-Jun-Yi/HTGNN.