Mathematical Finance and Machine Learning;
My primary research interests include risk measurement and management, portfolio optimization, quantitative finance, data analytics, and artificial intelligence. These fields present fascinating and challenging problems that draw on advanced mathematical knowledge and techniques. My recent and ongoing research projects focus on risk-sensitive reinforcement learning, optimal decision-making, portfolio management, risk assessment, and the application of machine learning in finance.
In a recent project, we proposed a novel learning algorithm designed to leverage historical data for improving the prediction accuracy of credit ratings. Specifically, we developed new metrics that can be applied to various artificial intelligence methods and introduced a new ensemble structure of artificial neural networks. Real-world data was then used to validate our proposed model for credit rating prediction through comprehensive performance measures.
In another project, we applied sequence-based clustering to group firms exhibiting similar transition behaviors. Credit risk assessment was conducted using representative transition matrices derived from these clusters. The proposed clustering model was evaluated across three long-term classification scenarios.
