About
Hi, I’m Eric.
I am a quantitative researcher focused on systematic equity investing, portfolio construction, and research infrastructure.
My work sits at the intersection of financial theory, statistics, machine learning, and software engineering. I am interested in one practical question:
How do investment ideas become investable portfolios?
That question spans the entire research process — from raw data and factor design to prediction models, portfolio construction, and realistic evaluation under execution constraints.
I care less about maximizing backtest performance and more about understanding whether a signal remains robust, explainable, and deployable in live investment environments.
I am currently based in Shanghai and work at Systematica Investments on equity alpha research within systematic investment strategies.
Research Interests
My research interests include:
- Equity alpha research and factor investing
- Signal generation and predictive modeling
- Portfolio construction and optimization
- Robustness and out-of-sample validation
- Financial data infrastructure
- Research systems and reproducible workflows
Building BagelQuant
Alongside my professional work, I am building BagelQuant — an open-source ecosystem and knowledge platform for quantitative equity portfolio management.
BagelQuant aims to connect:
Data → Factor → Prediction → Portfolio → Backtest
The goal is to make quantitative research:
- composable rather than monolithic
- transparent rather than hidden behind abstractions
- rigorous without sacrificing iteration speed
The software layer is designed around computation graphs, where datasets, transformations, signals, forecasts, and portfolio logic become reusable building blocks that can be assembled like Lego.
BagelQuant also serves as a public knowledge base covering topics across quantitative finance, mathematics, machine learning, optimization, and research engineering.
Research Principles
Investability over Backtests
A strong historical result is not enough.
I focus on whether ideas survive realistic assumptions around turnover, costs, portfolio constraints, and changing market regimes.
Explain Before Optimize
Models should help understanding before improving metrics.
Interpretability and structural intuition matter.
Make Assumptions Visible
Research systems should expose dependencies, transformations, and decisions instead of hiding them.
Experience
Quantitative Researcher— Systematica Investments
Shanghai, China
- Equity alpha research for systematic investment strategies
- Signal evaluation and robustness analysis
- Portfolio-oriented research under production constraints
Selected Research
Analyst–Management Disagreement
Research assistant project using NLP on earnings calls to study disagreement and subsequent market reactions.
Volatility Forecasting & Risk Timing
Applied GARCH and machine learning methods for volatility prediction and dynamic portfolio risk allocation.
Open-Source Quant Research Tools
Building tools and infrastructure including (link to GitHub repos):
Technical Stack
Python · SQL · C++ Data Engineering · Quant Research Infrastructure · Financial Modeling
Outside Work
Outside research, I enjoy building systems, writing technical notes, and exploring how software architecture can improve the investment research process.
Connect
- Email: [email protected]
- LinkedIn: linkedin.com/in/yanzhonghuang
- GitHub: github.com/bagelquant
- Website: bagelquant.com