Calculation of default probability of companies based on structural models by considering default correlation

The purpose of this research study is to predict the default probability of automobile and food industries in the corporate banking portfolio of banks, (not) considering the correlation of default. In this study, 30 companies from the automobile industry and 30 companies from the food industry whose stock market information is available were randomly selected. The market information of these companies was used daily from August 1, 2018 to August 31, 2019 for modeling purposes. The modeling of default probability of companies was done based on estimation of asset value, asset volatility, and drift rate. Multivariable GARCH (MGARCH) model was used to estimate the parameters of CAPM model, which is necessary to predict the drift rate. Through calculating the average default probability of companies in each industry, the probability of default of that industry (without considering the correlation of default) was obtained. Also, the asset value approach was used to calculate the default probability of each industry by considering the default correlation. In the asset value approach, maximum likelihood method was used to estimate default correlation parameters. The results showed that the prediction of the probability of default of the selected industries by considering the correlation of default is more consistent with the actual default.