true positives labA reproduction of the consensus feature-selection workflow from Sagaceta-Mejía (2024), Predicting Stock Market Movements in Emerging Markets Using Optimized Technical Indicators and Neural Networks.
It runs on ECH / EWZ / IVV across 216 technical indicators: consensus feature selection, MLP training, and the results — with notes on which tactics hold up well enough to reuse.
The full work lives in a Jupyter notebook with the data, models and charts: