School of Business and Management Department of Finance 178 Risk-Management Duel: Financial vs Operational Risk Management Supervisor: MACKAY Peter / FINA Student: BI Zihao / ECOF CHAN Wan Yu / GBUS HAN Chengru / QFIN KAN Robin / GBUS LAI Yuen Ying / ECOF LEUNG Hiu Hin / ECOF LO Hong Yi / FINA PANG Hoi Wun / QFIN TENG Josh Bennet Choi / BTBGM ZHANG Mengting / SBM Course: UROP 1000, Summer UROP 1000, Summer UROP 1000, Summer UROP 1000, Summer UROP 1000, Summer UROP 1000, Summer UROP 1100, Summer UROP 1000, Summer UROP 1100, Summer UROP 1000, Summer Companies typically mitigate risk through financial and operational risk management. This project aims to analyze the relationship between these two in evidence from mining companies. At the current stage, the research process focuses on operational risk management, studying how companies adjust their production and production capacity in response to changes in metal prices. We have collected data on various items related to the companies’ production levels in different locations and years, such as mineral reserves, mineral resources, and annual production. This report provides a summary of the data collection process in Summer 2024, highlighting specific findings and the challenges encountered for each company. Applications of Machine Learning to Financial Data Supervisor: NOH Don / FINA Student: NGUYEN Kim Hue Nam / COGBM YANG Zimo / ECOF Course: UROP 1000, Summer UROP 1000, Summer We replicated and evaluated the study presented by Jiang, Kelly, and Xiu (2023) that uses an image-based machine learning approach to predict future stock returns. We focused on experiments that use a 20-day price trend to predict returns over the next 5 days, recorded adjustments that differed from the original authors, and showcased portions of our replication code. The experimental results indicate the method’s predictive performance is encouraging, and we believe there is value in further validation after addressing certain technical constraints and using larger data volumes. Meanwhile, we explored the application of the method in the Korean stock market and made suggestions for subsequent transfer learning research.
RkJQdWJsaXNoZXIy NDk5Njg=