Credit Risk Assessment Of Listed Companies In Transportation Industry Based On BPSO-RF Model

The transportation industry is my country’s basic industry and service industry related to the national economy and people’s livelihood.It is an important sector of the national economy.It has a high degree of correlation with other industries and is a guarantee for the effective operation of other related industries.Because the transportation industry is cyclical,susceptible to factors such as government policies and pressure on capital expenditures,the transportation industry is an industry with special risks.Constructing a credit risk credit evaluation system for transportation enterprises will help enterprises to objectively understand their own credit risk status,improve risk management awareness,and promote the healthy and sustainable development of transportation enterprises.This article uses the transportation listed companies on the Fonghuashun Transportation Board and the New Third Board in each quarter from 2015 to 2020 as the research samples,combined with the characteristics of the transportation industry,and builds the credit of transportation companies from the financial and non-financial aspects of the company.Risk indicator system;use the KNN method to fill in the missing values,and use the K-means algorithm to down-sample to process the unbalanced data;on this basis,through the initialization of the particle swarm,each individual in the particle swarm is binary coded,using discrete Binary particle swarm optimization algorithm is used to optimize the number of decision trees in random forest and the maximum depth of decision trees,and build a credit risk evaluation model for listed transportation companies based on the BPSO-RF model.A comparative analysis of the prediction effects of multiple models based on the accuracy rate,recall rate and other criteria shows that the prediction accuracy rate of the BPSO-RF model constructed in this paper is up to 91.40%,and the random forest model predicts on the same verification set The correct rate is only 83.87%,the AUC value of the BPSO-RF model is 0.913,and the AUC value of the random forest model is 0.837.The random forest models are all greater than the prediction effects of the SVM model,the LR model,and the DT model.The main conclusions of this article are:First,in the transportation sub-industries,the port transportation industry(0.756),logistics transportation industry(0.743)and water transportation industry(0.735)ranked the top three,with credit scores exceeding 0.7 points,road transportation industry(0.682),railway transportation industry The industry(0.677)and the air transport industry(0.654)rank the bottom three.In 2020,due to the impact of the world’s new crown epidemic,the total transportation turnover,passenger turnover and passenger throughput of the air transport industry will drop by 38.3%,46.1% and 36.6% year-on-year.Poor operating profitability and solvency will affect the industry’s credit The main reason for the low risk score.Although port transportation companies are also more severely affected by the world’s new crown epidemic,in the second half of 2020,due to the proper control of the epidemic in China,the global demand for Chinese materials is strong,resulting in a rare peak period for freight growth in the container market.This is its profitability and compensation.The main reason for the increase in debt capacity.Second,among the bottom 20 transportation companies with low scores,5 are in the logistics and road transportation industries,4 are in the railway and air transportation industries,and 2 are in the port transportation industry.The financial indicators of the penultimate ST Long Investment and the penultimate Jiangxi Changyun are far below the industry average.The poor solvency and profitability of the two are the main reasons for the low credit scores,and the third is the bottom ~*ST Antong has a low credit score due to its credit history and poor profitability,but the company’s growth ability is performing well,which is the main reason why companies apply to the Shanghai Stock Exchange to withdraw the stock trading delisting risk warning.Third,among the top 20 transportation companies with higher scores,there are 13 in the logistics and transportation industry,and the third to 12 th with higher scores belong to the logistics and transportation industry;the railway transportation industry,air transportation industry and waterway transportation industry all belong to the logistics transportation industry.There are 2 stores.In 2020,under the situation of the new crown pneumonia epidemic,logistics supported the smooth operation of the entire social economy during the epidemic,linking the various systems of the entire society,and played an indispensable role.The epidemic has improved the status and value of logistics,and enabled logistics companies to improve their profitability.This is the main reason why they have many seats in the top 20 companies with better credit scores.Fourth,according to the analysis of the importance of the indicators,the asset-liability ratio,quick ratio and current ratio under the corporate solvency criteria are ranked 1,2 and 4respectively,indicating the main basis for the credit risk evaluation standards of transportation companies It is the company’s ability to repay debts;the company’s own basic quality standards and the number of undergraduates are ranked 3th and 6th respectively,ranking first,while the quality of corporate leaders ranks 22nd;corporate profitability The four indicators of gross margin of sales,total operating income and net sales margin under the standard level are all ranked in the top 10,indicating that the impact of corporate profitability on the credit risk evaluation results of transportation companies is second only to the corporate solvency.