Date of Publication :3rd July 2024

Abstract: The process of determining the possibility that a borrower would miss payments on a loan or meet a contractual commitment is known as credit risk analysis. For lenders, investors, and other financial institutions to make well-informed choices regarding loan extension or investment in a certain company, this study is essential. Analyzing credit risks and loan repayments is one of the biggest challenges that the modern world faces. There are a lot of defaulters in the world in different loan types and variations. According to a recent study conducted by CNBC in 2023 stated that there is an increase in the percentage of defaulters in India to 32.9%. The financial stability report (FSR) of the Reserve Bank of India (RBI) states that the gross non-performing assets (NPA) of public sector lenders in the credit card category was 18%, whilst private sector banks recorded a GNPA of 1.9 percent in FY23. According to a recent report from S&P Global the loan defaults in the U.S.A markets can rise to 3% by September 2024. This creates a demand for tools which can help big banks to grant loans to individuals or companies who have a good credit score. This research paper aims at providing an answer to the question of which machine learning algorithm will be best to perform such kind of predictions and can be used in the future by different credit risk analysis tools.

Reference :