An Ensembling Architecture Incorporating Machine Learning Models and Genetic Algorithm Optimization for Forex Trading

Algorithmic trading has become the standard in financial market. Traditionally, most algorithms have relied on rule-based expert systems which are a set of complex if/then rules that need to be updated manually changing market conditions. Machine learning (ML) is natural next step algorithmic because it can directly learn patterns and behaviors from historical data factor this into decisions. In paper, complete end-to-end system proposed for automated low-frequency quantitative foreign exchange (Forex) markets. The utilizes several State Art (SOTA) machine strategies combined under an ensemble model derive signal trading. Genetic Algorithm (GA) used optimize maximizing profits. also includes money management strategy mitigate risk back-testing framework evaluate performance. models were trained EUR–USD pair Forex Jan 2006 Dec 2019, subsequently evaluated unseen samples 2020 2020. performance promising ideal achieved about 10% nett P&L with −0.7% drawdown level based data. Further work required calibrate costs & execution slippage real It concluded increased volatility due global pandemic, momentum behind adapt environment will even stronger.