Adaptive Market Making Strategy: Extended Research on the Avellaneda-Stoikov Framework Based on Machine Learning

Introduction / Main Floor: Adaptive Market Making Strategy: Extended Research on the Avellaneda-Stoikov Framework Based on Machine Learning

This is a quantitative finance research project that explores how to integrate short-term volatility and direction prediction into the Avellaneda-Stoikov market making framework. It uses the XGBoost model and AAPL high-frequency limit order book data for backtesting, and compares the performance of the benchmark strategy with the machine learning-enhanced strategy.