基于HPO-SVC的房地产企业信用风险预测

打开文本图片集

中图分类号：F299.233.4

CreditRiskPredictionofRealEstate Enterprises Basedon HPO-SVC

YANG Zhi-hao

(School of Mathematics and Computer Applications, Shangluo University,Shangluo 726ooo, Shaanxi)

Abstract:As to the credit risk of liste dreal estate enterprises，acombined credit risk prediction model integrating HPO-SVC with SMOTEENN and RFECV is proposed.This model eliminates the randomness in parameter selection for the SVC classifier and addresses issues such as data missingness,class imbalance, and high dimensional redundancy of indicators in the credit risk indicator system.Taking 117 listed real estate enterprises from 2O18 to 2O22 as the research objects,comparative experiments are designed to verify the model'seffctiveness.The results show that theconstructed combined credit risk prediction model for listed real estate companies exhibits superior performance and higher classification accuracy.

Key Words:real estate enterprises;credit risk forecast; unbalanced data; support vector machines; hunter preyoptimization

房地产行业因与众多上下游产业相关联，曾是经济增长的重要动力，但其“高杠杆、高负债、高周转"的传统发展模式已不可持续。（剩余15824字）