Investigating the Factors Affecting the Stock Price crash risk with Emphasis on Feature selection Attitude

The main purpose of this study is to investigate the factors affecting the stock price crash risk using the feature selection approach in companies listed on the Tehran Stock Exchange. In this research, the data of ۱۰۱ companies in the period ۲۰۱۵-۲۰۲۰ have been studied. In order to select the optimal variables in each of the three criteria for calculating the stock price crash risk that have been used in this study, the method of selecting the cover feature has been used, Therefore, ۱۹ independent variables were entered into the model as input to the particle cumulative algorithm Finally, in each of the criteria of negative skewness, maximum sigma and Swing, a number of optimal variables were selected. Then, using the STATA software, the effect of various factors on the stock price Crash Risk was investigated. The results indicate that there is a positive and significant relationship between market value to book value and the stock price Crash Risk in all three criteria. But there is no significant relationship between firm size and the stock price crash risk in low-to-high swing and maximum sigma criteria. Also, working capital management and the stock price crash risk in the criteria of negative skewness, maximum sigma and low to high swing are not significantly related. While the Stock trading volume in all three criteria has a positive and significant relationship with the stock price crash risk. Taxes in both the criteria of negative skewness and swing have a positive and significant relationship with the stock price crash risk. While there is no significant relationship between the independence of board members and the stock price crash risk. There is no significant relationship between the lack of transparency of financial information and the stock price crash risk in the negative skewness and sigma criteria.

Keywords:

feature selection , stock price crash risk , Cumulative particle algorithm