Credit Risk Measure of Listed Pharmaceutical and Biological Companies Based on Genetic Algorithm KMV Model
Abstract
In order to measure the credit risk of listed companies in China's pharmaceutical and biological industry, a total of 28 listed companies in the A-share ST category and non-ST category were therefore selected as samples, and the model was improved using genetic algorithms, while the credit risk of 290 listed companies in the A-share pharmaceutical and biological industry in 2019-2021 was analyzed based on the improved model, and the research results showed that: the improved KMV model can effectively identify the improved KMV model can effectively identify the credit risk of listed companies in the industry, and the accuracy of the improved KMV model in determining whether an enterprise is in default reaches 78.57%; the credit risk of the pharmaceutical and biological industry decreases in the year of the outbreak of the new crown epidemic in 2020, and increases and the credit risk of enterprises appears polarized one year after the outbreak of the epidemic.
Keywords
Download Options
Introduction
Since the outbreak of the new crown epidemic in 2020, the public's health protection needs have increased and medical supplies for daily protection will be consumed in large quantities. For example, medical supplies such as masks, Lianhua Qingpu and nucleic acid reagents will be consumed dramatically after the epidemic, and the pharmaceutical industry is getting more and more attention. Based on the above reasons, this paper tries to measure and analyze the credit risk of listed pharmaceutical manufacturing enterprises in the western region of China, so as to promote this type of enterprises to be able to prevent their own credit risk more effectively and improve their credit quality.
Among the models for quantitative analysis of credit risk, the KMV model has the advantages of easy access to data, more intuitive calculation method, and better fit between the calculation results and the real credit risk changes. Yingchun Liu and Xiao Liu [1] (2011) used GARCH (1,1) volatility model to estimate the equity value volatility, and applied the KMV model to calculate the three-year default distance and expected default probability of 16 listed companies. The results show that the KMV model can well discriminate the difference in credit risk between ST and non-ST companies. Tang, Zhenpeng [2] and other scholars (2016) selected three industries to form the research sample, applied the modified KMV model to measure credit risk, and used factor analysis to elaborate the financial factors affecting the credit risk of listed companies in different industries. The research results show that the credit level of listed companies in the eastern coastal region is the best, followed by the central region and the western region is the worst; the credit risk of listed companies in the real estate industry is the least, followed by the pharmaceutical manufacturing industry.
Jinghai Feng[3] et al. redefined the optimal default point in the classical KMV model by combining genetic algorithm. The results show that the improved model fitting results show that the improved model fitting is more correct than the original model, i.e., the improved KMV model is more suitable for the creditworthiness assessment of listed companies in China. Shuyuan Zhou[4] used GARCH(1, 1) model to optimize the volatility and genetic algorithm to optimize the default point coefficients to modify the KMV model, and the results show that the GA-GARCH-KMV model is more accurate for the risk measure of listed companies with overcapacity industry.
Conclusion
This paper uses the basic principles and calculation methods of KMV model and genetic algorithm to improve the default point parameters in the traditional KMV model by selecting pharmaceutical and biological listed companies marked as ST in the last three years and companies in the same industry with comparable asset value with their assets, and using MATLAB software to build a GA-KMV model to make it more consistent with the actual situation of the Chinese A-share market. Finally, the modified model is applied to empirically analyze the credit risk profile of Chinese A-share listed pharmaceutical and biological companies, and the conclusions are as follows:
Using the GA-KMV model to modify the parameters of 28 sample enterprises, the corrected default point α value is 0.318 and β value is 3.517. The comprehensive accuracy of the modified model for judging whether an enterprise is in default is 78.57%, which can effectively distinguish the credit risk of ST class enterprises and non-ST class enterprises, but if the traditional KMV model is used, its comprehensive correct rate of default or not However, if the traditional KMV model is used, the combined correct rate of default and non-default is only 53.57%, i.e., it cannot effectively distinguish the difference between ST class and non-ST class enterprises.
Using the improved KMV model to measure the credit risk of 290 listed companies in China's A-share pharmaceutical and biological industry in 2019-2021, the analysis shows that the credit risk of the pharmaceutical and biological industry decreased and then increased in the past three years, and the credit risk of the pharmaceutical and biological industry decreased in the year of the outbreak of the new crown epidemic in 2020, and increased one year after the outbreak of the epidemic, and the credit risk of the industry was polarized phenomenon. In the future, the concentration of the pharmaceutical and biological industry will probably continue to increase, and there is more room for the development of head enterprises.