Finance Researcher
Asset Pricing · Derivatives · Machine Learning
I’m Henry, a quant graduate from Carnegie Mellon University. I have a strong background in finance and computer science, with experience spanning empirical financial research, quantitative modeling, and applied machine learning. I was nominated as one of the four Qatar Campus Scholar of Class of 2026. My research interests include asset pricing and derivatives. Beyond academic research, I have worked on portfolio management and asset allocation models and developed AI-driven investment analysis tools, using Python, R, and SQL.
Education
Carnegie Mellon University
Bachelor of Science in Business Administration
Concentration: Finance (Quantitative) Minor: Computer Science
Honors:
Qatar Campus Scholar (2025) Awarded to Top four graduate for academic excellence and leadership
University Honors (2025) Distinguish Academic Achievement
Past Papers
The Predictive Power of Options: Implied Volatility and Volume Dynamics Around Corporate Earnings
Author: Phat Diep
Advisors: Veli Safak, Serkan Akguc
Institution: Carnegie Mellon University in Qatar
This paper examines whether pre-earnings option market signals predict post-earnings stock returns, focusing on implied volatility and trading volume dynamics.
When the Government Closes: Economic Responses to Federal Shutdowns
Author: Phat Diep
Institution: Carnegie Mellon University in Qatar
This paper examines the immediate economic effects of U.S. federal government shutdowns using high-frequency data from the Weekly Economic Index (WEI).
Projects
The AI-Driven Investment Recommendation System is a Streamlit-based application that gathers a detailed customer investment profile and processes it through multiple AI agents to generate:
- Personalized portfolio recommendations
- Quant-driven performance expectations
- Full risk assessment
- Backtesting visualizations
- Shariah compliance assessment
- Narrative suitability reports
- Sector allocation insights
Poster: drive.google.com/AI-Portfolio-Advisor
- Implemented OpenCV preprocessing pipeline for robust face detection and normalization
- Applied advanced regularization techniques including dropout and batch normalization
- Designed CNN architecture using TensorFlow + Keras, achieving 94% classification accuracy