Quantitative Research Divison
Algorithm Implementation of SOTA Online Portfolio Selection
Contributor: Victor Xiao, Dustin Yu, Yanpeng Wang, John Zhou, Polo Li, Joanne Wu, Kevin Wang, Jason JiOnline portfolio selection is a fundamental problem in computational finance, which has been extensively studied across industries and research communities. Such, we constructed a systematic methodology for online portfolio selection algorithms, including benchmark, follow-the-winner, follow-the-loser, pattern-matching algorithms and meta-learning augmentation. The project aims to provide a comprehensive survey and a structural understanding of published online portfolio selection techniques. The project is currently open-sourced here.
Multi-factor Portfolio Strategies Framework
Contributor: Polo Li, Victor Xiao, John Zhou, Troy Wu, Jason Ji, Hankle Liu
State-of-the-art Multi-factor investment framework, constituting factor mining, data cleaning, factor analysis, factor integration, and portfolio optimization modules. The universe in which factors are mined and explored can be divided into three features space: fundamental factors, high-frequency statistical factors(quantity-price factor) and macro-economic factors.
Transformer with Hierarchical Risk Parity Embeddings [ETC x Risklab Colaboration]
Contributor: Johnson Zhao, Hankle Liu, John Zhou
Through utilization of Natural Language Processing Techniques, mainly, Transformer encoders with embedding of financial time series such as log return and volatility of stocks and industry index into learned representation space; the project explored a state-of-the-art approach of systematic predication on the financial market.
Multifactor Strategy Backtesting with Python/R
Contributor: Jack Chen