The Application of Long Short-Term Memory (LSTM) for Stock Prediction, Anomaly Detection, and Risk Analysis in the Banking Sector

Stock market prediction and anomaly detection are crucial for making accurate investment decisions, especially with technological advances that enable more precise data analysis. The method used is Long Short-Term Memory (LSTM). In this study, anomaly detection will be conducted to identify unusual movements in stock price fluctuations. Subsequently, based on the periods of stock price movements identified as anomalies, a risk analysis will be performed on BBRI, BBCA, and BMRI stocks, this can help investors make better decisions about investing in BBRI, BBCA, or BMRI companies. Then, a prediction analysis on BBRI, BBCA, and BMRI stock prices will be conducted using the LSTM algorithm. The research findings provide recommendations for investing in more profitable companies. In this study, BBCA outperformed BBRI and BMRI as there were no anomalies and it had lower risk analysis results.