Blog
#machine-learning
A Scalable ML Framework with Monadic Design
A Python ML framework built on monadic design: DataPod state carriers, composable trained transformers, and reproducible pipelines from research to production.
Markov Chains for Risk Model Labelling and Portfolios
Markov Chains turn binary default labels into continuous default probabilities, improving risk models on small data and enabling fairer portfolio comparisons.
LSTM and CNN Models for Crude Oil Trading
Hybrid CNN and LSTM models beat ARIMA, RSI, and MACD on WTI crude: LSTM-CNN reached 17.8% average annual ROI with the best Sharpe ratio we tested.
