Corporate performance: SMEs performance prediction using the decision tree and random forest models

Abstract

Stock markets are volatile and continue to alter based on the functioning of the company, historical documents, market-rate, and news updates with the timings. Stock price prediction is the utmost stimulating assignment. In the present communication, a study with data on the stock prices of the top small and medium-sized enterprises (SMEs) in the National Stock Exchange of India (NSE) was utilized to estimate the functioning of the technique executed. The results of this study demonstrate the impact of COVID-19 on the financial distress of SMEs and also helps us in understanding how a better prediction model can help in predicting financial distress. Many studies have been conducted to estimate the bankruptcy of the SME sector using accounting-based financial. But in this study, the leading principle was to exemplify the means to utilize machine learning (ML) algorithms in the bankruptcy prediction of SMEs. The outcomes from the proposed a decision tree and a random forest prototype are observed to be effective with a high accuracy rate. The study has practical implications on the prediction accuracy and practical value for banks in supporting the financial decision and can be used to access the loan applications of SMEs.