Investigating of factors affecting conservatism with emphasis on feature selection attitude

The issue of feature selection is one of the most important issues in machine learning, which is very important in many applications because in these applications there are many features, many of which are either unused or do not have much information load. Failure to remove these features will increase the computational load. This is also important in financial matters. One of the most important issues among stock exchange companies in Iran is conservatism, which is influenced by many factors. Therefore, identifying the optimal variables, in this case, is very important. This study aims to identify the optimal variables affecting conservatism using a particle aggregation algorithm as a feature selection method. The results show that the variables of financial leverage, the ratio of market value to book value, rate of return on assets, and return on equity have been selected as the optimal variables.

Keywords:

feature selection , particle aggregation algorithm , conservatism