Quant
Exploring the market's hardest problems.
Quant Team
The Quant Team focuses on the quantitative aspects of finance. We critically engage with financial markets by exploring statistical models, derivatives pricing, and algorithmic trading strategies. Members develop coding and data analysis skills to create and test financial models on market data.
Key Activities
- Regular Learning Sessions
- Playing Market Games
- Discussing Brainteasers and Puzzles
- Participation in Quant Competitions
Our Projects
This Semester
In the current semester, the Quant Team continues with two hands-on projects that combine financial modeling, data analysis, and software engineering.
Kaggle Stock Return Fundamentals
Members work on a Kaggle-style finance challenge that predicts one-year US stock returns from SEC-filing fundamentals. The project focuses on leakage-safe validation, robust tabular models, feature engineering, and disciplined experiment tracking.
Project Goals
- Build reliable baselines for noisy cross-sectional return prediction
- Analyze fundamentals, missing values, outliers, and sector effects
- Create reproducible submissions and compare model improvements
C++ Backtesting Engine
The team develops a high-performance, event-driven backtesting system for quantitative trading strategies in C++20. The project combines quant research with systems engineering, performance work, and clean software design.
Project Goals
- Implement and test core backtesting components such as data handling, portfolios, strategies, and execution
- Improve performance with cache-friendly data structures and zero-allocation hot paths
- Prepare the engine for multi-asset research, benchmarks, and future dashboard integrations
Quant Competition WS 25/26
As one of our key projects for this semester, the Quant Team hosts its own internal Quant Competition, open exclusively to members of our group. By joining our group, you can participate individually or in pairs of two. No prior knowledge of financial markets or trading algorithms is required. Everything you need to know will be taught during our weekly quant sessions.
How it works
In the competition, you'll receive synthetic price data and your goal will be to develop a trading algorithm that exploits patterns in the data. After the kickoff, we'll periodically release new time series for you to use as training sets, giving you 2 to 5 weeks between rounds to refine your strategy. On the evaluation date, your algorithm will be tested on unseen data, and the team achieving the highest Sharpe ratio in the final evaluation will be crowned the winner. The top 3 teams will receive prices.
Getting Started
Clone our GitHub repository and follow the instructions from the readme file. Keep in mind you'll be only able to make a submission with valid credentials that will be handed out to you when you registered your team name.
Key Dates
Kickoff
Evaluation 1
Evaluation 2
Evaluation 3 (Final)
Organized by Jonathan Willert, Mathis Makarski, and Hendrik Scherer.
Regular Learning Sessions
During the semester, Quant Team members will select a topic to present in our regular meetings in teams of two. The presentations can be in any format of your choice like slides or plots from a jupyter notebook. These learning sessions are held every 1–2 weeks, either in person or in a hybrid format, and attendance is expected from all members. This is your opportunity to dive deep into a topic of your choice and share your insights with the rest of the group. No prior knowledge is required, members from all backgrounds and experience levels are welcome. Teams will be formed during the kickoff, and throughout the semester, you'll have time to research, learn, and prepare a presentation that helps the entire group grow.
Previous Topics
- Mean-Variance Portfolio Optimization
- Fundamental Factor Models
- Options Pricing with Black-Scholes
- Risk Management Techniques