Weiminghui Ji

Quantitative Researcher

Email | LinkedIn | GitHub

About Me

Weiminghui Ji is a quantitative researcher specializing in interest rates, finding joy in using systematic, interpretable methods to solve real-world financial problems. Currently, she works at JPMorgan Rates Flow as a desk quant, with her experience also extending to linear instruments and exotic derivatives. Diligence is engraved in her DNA 💪

Experience

JPMorgan Chase

Quantitative Research Associate, Rates Options Feb 2022 - Present
  • SOFR Futures Arbitrage: Built an arbitrage strategy to trade quarterly SOFR futures against the monthlies with Python. Conducted thorough analysis of FOMC risk and granular SOFR refix risk. Constructed an efficient frontier to trade off expected PnL against the risk undertaken. Implemented continuous enhancements to the computational efficiency to support live trading.
  • TermSOFR Refix Hedge: Developed a stepwise SOFR curve with VWAP of futures in accordance with CME TermSOFR methodology. Solved for optimal future weights to hedge TermSOFR refix risk through regression across various FOMC scenarios. Backtested hedging effectiveness and optimized the timing of hedging, focusing on month end with TWAP as a benchmark.
  • Vol-of-Vol and Skew Relative Value: Created a methodology to imply swaption vol-of-vol and skew parameters from strangles and risk-reversals. Implemented the realized version of these parameters and automated daily generation of relative-value reports for actively traded structures.
  • TBA Option: Implemented implied and realized hedge ratios for TBA Options against various instruments including treasury future options, bond forward options, and swaptions. Conducted rolling regression analysis of TBA realized volatility against FV, TY, and US realized volatilities with forward feature selection.
  • Bermudan Option: Programmatically selected high-quality benchmark instruments from Totem Data. Developed a volatility adjustment curve based on the expected lifetime. Built an arbitrage detection tool for Accreters with Bullet prices adjusted to Totem data.
  • Pricing Library: Supported pricing of SOFR caps by adjusting time-to-expiry, life-cycling of LIBOR fallback vanilla options and Bermudans with C++, quick parsing and pricing of YCSO and forward-vol queries.
Quantitative Research Summer Associate, Rates Options Jun 2021 - Sep 2021
  • Swaption Backtesting: Designed a scalable framework for backtesting Gamma selling strategies with various Delta hedge ratios. Developed customized backtesting tools based on relative values such like implied/realized vol.

Education

Columbia University

M.S. in Financial Engineering
2020 - 2021

Peking University

B.S. in Data Science
2016 - 2020

Skills

Programming
Python - C++ - SQL - Hive - Scrapy
Data Science
Regression Models - Decision Tree - Deep Neural Networks

Languages

Chinese - Native

English - Professional

Weiminghui Ji

Quantitative Researcher

Email | LinkedIn | GitHub

Experience (Cont.)

Guotai Junan Futures

Quantitative Research Intern, Asset Management Department Jun 2020 - Sep 2020
  • Enhanced CTA Strategy: Developed Python-based backtesting framework for futures CTA strategies with contract rolling. Backtested 30 commodity futures over 7 years using 5-minute data for hyperparameter optimization based on Calmar and Sharpe ratio.
  • Market Sentiment Feature: Identified and integrated a news click data source for tracking market sentiment by strategically parsing web source code. Utilized Scrapy to crawl commodity news data and applied natural language processing techniques to generate alternative features. Backtested the daily close data and iterated on the realtime API.

Yinhua Fund Management

Quantitative Research Intern, Strategy Development Department Jan 2019 - Apr 2019
  • Financial Text Analysis: Applied TextRank algorithm to generate financial news summary with Python.
  • Deep Learning & Natural Language Processing: Researched relationship between financial news and the stock market, trained a regression model to predict stock returns using TFIDF features and sentiment features extracted by LSTM model.

Fosun

Data Analyst Intern, Algorithm Group Jan 2018 - Feb 2018
  • Insurance Claim Prediction: Developed user portraits using SQL on Hive with car insurance data. Trained Gradient Boosting Decision Trees using XGBoost and k-folds cross-validation, achieving a 90% F1 score for risk prediction on training data. Successfully delivered and validated the model with the car insurance company.

Awards

Outstanding Graduate, Peking University

2020

China National Scholarship

2019