IMC Trading Challenge Bots

Overview

Implemented rule based and statistical strategies for simulated products, balancing inventory risk and PnL.

Developed for the Algothon 2025 competition, this project involved building a trading strategy algorithm to perform optimally given certain metrics (mean(PL) - 0.1 * StdDev(PL)). The solution involved assessing provided price data from a simulated trading universe of 50 instruments, building a predictive model, and back-testing it. The algorithm, implemented in Python, trades position differences based on the most recent price, managing a $10k position limit per stock. Key considerations included optimizing for trade frequency, projecting worst-case scenarios, and implementing risk minimization techniques.