Trading Strategy

This project is under quick development.

Adaptive Regime Strategy

ARS Intro. ARS is a finished experiment.

This is a Mixture of Experts agent consists of:

Trend Agent: Transformer framework aiming to profit when the ticker is surging
MeanRev Agent: Mean Reversion Strategy targetting chopping
Crisis Agent: exclusively for saving the agent during bear market. This agent is extremely effective, gaining an alpha of 20% on GOOG from 2024 to 2025.

Name		Name	Last commit message	Last commit date
Latest commit History 81 Commits
Architecture_v1		Architecture_v1
Archived		Archived
Plots		Plots
__pycache__		__pycache__
.gitignore		.gitignore
LICENSE		LICENSE
Multi-Agent Trading Strategy Framework.md		Multi-Agent Trading Strategy Framework.md
README.md		README.md
agents.py		agents.py
data_utils.py		data_utils.py
feature_engineering.py		feature_engineering.py
regime_detector.py		regime_detector.py
v13_moe_model.py		v13_moe_model.py
v13_multi_ticker_test.py		v13_multi_ticker_test.py