基于机器学习的债务风险管理体系构建与应用——以浙江省交通投资集团为例

Construction and Application of Debt Risk Management System Based on Machine Learning:An Example of Zhejiang Communications Investment Group

2021年2月,国务院国资委印发了《关于加强地方国有企业债务风险管控工作的指导意见》,强调了对地方国有企业债务风险的管控工作要求。如何抓住地方国企债务风险的特点,建设一套契合度高、实用性强、可落地应用的债务风险预警体系,是地方国有企业债务风险管控工作的关键。首先,本文梳理债务风险管理体系理论研究中管理制度、指标设定、模型构建的理论基础;其次,提炼债务风险管理体系的主要建设目标,包括风险可衡量、成因可分析、执行可跟踪、隐患可预警、影响可控制五大管控目标;再次,提出有效实施债务风险管理体系的五项具体应用步骤,助力集团债务风险管理精度与质量双提升。

February 2021,the State-owned Assets Supervision and Administration Commission of the State Council issued "the Guiding Opinions on Strengthening the Debt Risk Management and Control of Local Stateowned Enterprises",emphasizing the requirements for the debt risk management and control of local state-owned enterprises.How to grasp the characteristics of the debt risk of local state-owned enterprises and build a set of debt risk early warning system with high fit,strong practicability and practical application is the key to the debt risk management and control of local state-owned enterprises.This paper firstly reviews the theoretical basis of management system,index setting and model construction in the theoretical research of debt risk management system.Secondly,the main construction objectives of the debt risk management system are refined,including five management and control objectives:measurable risks,analyzable causes,trackable implementation,early warning of hidden dangers and controllable impacts.Thirdly,this paper proposes five specific application steps for the effective implementation of the debt risk management system,which will help improve the accuracy and quality of the Group's debt risk management.