Stock Pledge Defaults Prediction Using Machine Learning

This study addresses the prediction of stock pledge defaults using machine learning techniques, a crucial topic in the field of economics and finance. Specifically, the study aims to develop models capable of identifying the risks associated with stock pledge defaults, thereby enabling investors and lenders to make informed financial decisions. The dataset used consists of 1,442 entries and 62 columns, including financial indicators such as pledge ratios, stock volatility, and returns. The results showed that using machine learning algorithms like K-Nearest Neighbors (KNN), Support Vector Classifier (SVC), and Decision Tree (DT) resulted in high accuracy in predicting defaults. These models demonstrated excellent performance, achieving accuracy rates of up to 99% in some cases, reflecting the ability of these techniques to handle complex financial data more effectively than traditional methods. This study contributes to the literature on financial risk prediction by providing an advanced framework that utilizes machine learning techniques, thus enhancing the understanding of the factors influencing stock pledge defaults. The study also highlights the importance of addressing data imbalance, opening new avenues for future research in this field.