Vivian (Jingcheng) Yu
About Me
I am a Ph.D. Candidate in Actuarial Science at the University of Waterloo,
specializing in the intersection of financial stability, risk management, and theoretical
mathematics. My research focuses on quantitative risk management and stochastic
optimization, developing robust frameworks that bridge abstract mathematical theory
with data-driven market applications.
With a Master of Financial Engineering from UCLA and professional experience at
Ernst & Young and the California Department of Insurance, I
combine rigorous academic discipline with industry-grade problem solving.
Education
University of Waterloo
UCLA
GPA 3.82 / 4.0
University of Manitoba
First Class Honor
Credentials
- ★FSA (Fellow of the Society of Actuaries)
- ★CERA (Chartered Enterprise Risk Analyst)
- ★CFA Level II Candidate
Technical Skills
Professional Experience
University of Waterloo — Teaching Assistant
Lead tutorials for undergraduate actuarial students, translating complex mathematical and risk-management theory into accessible, engaging concepts, with personalized mentorship during office hours.
Ernst & Young — Actuarial Consultant
Applied algorithmic optimization techniques (Simulated Annealing, Genetic Algorithms) to optimize portfolios with liabilities, successfully reducing financing costs by 50%.
California Department of Insurance — Actuarial Student Assistant
Conducted in-depth research on ISO filings and redeveloped the Class Plan Application using VBA, achieving a 30% reduction in processing time.
Research Projects
Submodular Risk Measures
Studies submodularity for law-invariant functionals: law-invariant coherent risk measures are submodular exactly when they are coherent distortion risk measures, including Expected Shortfall. Complete characterizations for shortfall risk measures via Arrow–Pratt risk aversion and for optimized certainty equivalents, with an empirical study on daily US equity returns.
With Ruodu Wang
READ ON ARXIV ↗ES-only E-backtesting for AR-GARCH Losses
Developed sequential e-backtests for Expected Shortfall when only the ES forecast is reported, treating the associated VaR as unobserved. Established e-process validity via infima over admissible values and certified threshold tests for AR-GARCH losses.
Joint work with Ruodu Wang
Mathematics of Scenario-based Risk Evaluation
Studies risk measures determined by collections of scenarios (probability measures), including scenario-based distortion and spectral risk measures and their structural properties.
Joint work with Ruodu Wang and Chenxi Xia
Portfolio Design with CART Models
Engineered a portfolio analysis framework using Classification and Decision Tree (CART) models across 14 asset classes and 73 macroeconomic variables. Identified critical shifts in feature importance across rolling windows.
Triggered Shorting Strategy on VIX
Led a team calibrating a 7-state Hidden Markov Model and K-means clustering to identify market regimes, feeding neural-network predictions that powered a short strategy with an average return of 173.5%.
Let's connect.
Open to research collaboration and conversations on risk, mathematics, and markets.