Chun Man CHAN

● Developed Python-based backtesting tools to analyze trading signals (e.g., moving averages, highs/lows, returns) for the top 1,000 U.S. stocks, automating daily stock selection and performance evaluation over multiple time horizons (1 day, 2 days, 1 month, etc.). ● Implemented Bayesian Optimization for constrained market signal search, optimizing combinations of factors to maximize average 5-day return, achieving faster and effective results compared to the original approach. ● Conducted research on mispricing and arbitrage opportunities in leveraged and inverse leveraged ETFs, identifying deviations from expected leverage ratios and assessing profitability potential.

● Conducted rigorous and comprehensive deep dives into quantitative methods, with a specific focus on the Markowitz Model, utilizing advanced analytical techniques to assess portfolio optimization and risk management strategies. ● Engaged in hands-on research, actively immersing myself in the subject matter and embracing the intricate research process, ensuring a thorough understanding of quantitative methods and their practical applications. ● Leveraged a multidisciplinary approach, integrating concepts from mathematics, statistics, and finance to develop robust models and frameworks for portfolio construction and asset allocation.

● Conducted in-depth analysis to identify the factors contributing to the company's customer increasing bankruptcy rate compared to the industry in the recent quarter. ● Utilized Python and SQL to aggregate, filter, and transform the unstructured customer data from different databases. ● Leveraged vintage analysis and data visualization techniques to uncover the key drivers like the abnormal increase in utilization and credit limit. ● Suggest a multifaceted approach for better risk management, including early credit line reduction and improvement on the machine learning model

● Researched and analyzed emerging trends in the technology industry, including advancements, market dynamics, and disruptive innovations, to support the development of Entrio's comprehensive database of technology companies. ● Conducted in-depth vendor research, evaluating technology providers, and assessing their products and services to identify key players and potential partners for inclusion in Entrio's database. ● Compiled and maintained Entrio's proprietary database, meticulously updating it with the latest information sourced from a diverse range of reliable and authoritative data sources, ensuring its accuracy, completeness, and relevance. ● Collaborated with cross-functional teams to gather and validate data, leveraging strong attention to detail and organizational skills to ensure the integrity and consistency of the database, enabling users to access up-to-date and insightful information on technology companies.

● Designed and implemented cutting-edge computer vision algorithms to empower the robot with the ability to autonomously navigate and execute complex tasks in diverse and challenging environments, leveraging image processing, object detection, and pattern recognition techniques. ● Developed and maintained robust and scalable software solutions to control the robot's movements and actions, ensuring seamless integration between the computer vision system and the overall robotic architecture. ● Collaborated closely with cross-functional teams, including mechanical and electrical engineers, to seamlessly integrate the computer vision system with the hardware components of the robot, enabling cohesive and synchronized operation. ● Conducted meticulous experiments and iterative testing to continuously enhance the performance and accuracy of the computer vision system, applying techniques such as data augmentation, model optimization, and algorithm refinement to achieve superior results.

Topic : Building knowledge graphs for suggesting machine learning models via LLMs. ● A HKUST program designed to provide undergraduate students with an opportunity to engage in academic research under the supervision of professors and PHD students. ● Conduct weekly or biweekly internal presentations based on research paper readings. ● Conduct data analysis and implementing algorithms.