Symbolic Fourier Approximation in MQL5: a Fourier front end with MCB binning for historical analog search, measured head to head against SAX on identical windows.
The repo hosts Bag-of-SFA-Symbols classifier from scratch in pure MQL5, turning price windows into words and bags of words into a regime classifier. Benchmarked against Dynamic Time Warping on BTCUSD, the BOSS ensemble wins on clean accuracy and runs about twenty times faster.