This is an evolving collection of trading models where I continuously build and refine machine learning pipelines for market prediction.
Each folder is a self-contained strategy with modular code and full backtesting support.
MSFT_algo/
- Gradient Boosting classifier trained on engineered MSFT financial features; full model validation using time-series CV and backtested trading logic to assess deployabilitySVM_pipeline/
– SVM classifier with macro features and time-series CV (tuned for tech stocks)x4_pipeline/
– 4-model ensemble (LR, RF, XGBoost, LGBM); fixed parameters used to avoid overfitting from excessive tuning
Each folder includes its own README.md
and requirements.txt
. Run independently.