Overview
What this challenge is about.
Simulate price paths for a single underlying (geometric Brownian motion is fine as a baseline; bonus for stochastic volatility). Implement Longstaff-Schwartz Monte Carlo as the classical baseline for American option pricing. Then train a feed-forward neural network that maps state (spot, time, current holding) to hedge ratios, minimizing terminal P&L variance plus a transaction-cost penalty. Compare hedging error distributions, sensitivity to transaction costs, and runtime. Write a 2-page commentary on when deep hedging beats the baseline and when it does not. Cite at least two papers and acknowledge the assumptions you made.
The Brief
What you'll do, and what you'll demonstrate.
Build a reference notebook comparing deep hedging to Longstaff-Schwartz Monte Carlo on a vanilla American put with realistic transaction costs.
Earning criteria — what you'll demonstrate
- Implement Longstaff-Schwartz Monte Carlo for American option pricing
- Train a deep hedger with a variance-plus-cost objective
- Design a fair comparison between a classical and a learning-based method
- Write clean research commentary new colleagues can learn from
Program Fit
Where this fits in your program.
Sharpens the same skills your degree expects you to demonstrate.
Skills
Skills you'll demonstrate.
Each one shows up on your verified credential.
Careers
Roles this prepares you for.
Real titles. Real skill bridges. Pick the one closest to your trajectory.
ML Researcher
Reproducing a paper and writing honest commentary on its limits is the literal day-one task for an ML researcher inside a quant fund.
This challenge sharpens
- deep-learning
- research-writing
- monte-carlo-methods
Research Scientist
Designing a fair comparison and grounding it in primary literature is exactly the rigor expected from a junior research scientist's first internal review.
This challenge sharpens
- stochastic-modeling
- derivatives-pricing
- research-writing