Marsto is a web app that adapts Dayder to find positive market correlations in over 12,000,000+ stock and ETF time-series records. It uses a custom binary serialization method to convert CSV files containing stocks and ETFs into a file format that is magnitudes faster and smaller than the original. The datasets used for this instance of the web app was provided by Boris Marjanovic and can be found on Kaggle.
All credit goes to Tristan Hume and Marc Mailhot for developing the Dayder project and credit goes to Boris Marjanovic for providing the datasets used in this project.
This project is released under the MIT license, see the LICENSE file for details.