Small transformer model for stock market intraday forecasting

Based on the concepts in "CIMTR" and others, swing trading

In this project, I generated investing insights by applying sentiment analysis on financial news headlines from Finviz.

Liferay 7 plataform to add a custom product to host trading strategies

This repository contains code, information, and resources developed and used for the Stevens Algorithmic Trading Competition Spring 2019. For questions about the competition itself, please contact algotrading@stevens.edu.

OHLCV Ethereum data is transformed using an encoder-decoder network. The transformed data is input into an LSTM model that predicts the asset price in 24 hours. These forecasts are used to create buy and sell signals which were back-tested producing a Sharpe ratio of 1.45. The code was run various times end-to-end producing similar results.

Rainflow‑aware DQN for battery dispatch (arbitrage + FR/reserves): Gymnasium env with safety layer, exact cycle‑based degradation, time features & stacked observations; SB3 training.

🔋 Optimize battery dispatch using a rainflow-aware DQN that tracks degradation and enhances performance with advanced features for reliable results.