End-to-end ML system for prediction market trading — 521K markets, 78 features, 7 model architectures, walk-forward validation, live VPS A/B across 7 configs. Honest research-stop on alpha decay (NO-GO verdict). AFML methodology: Purged K-Fold, Deflated Sharpe Ratio, meta-labeling, focal loss.

Multi term Polynomial Regression with Learnable Exponents and Coefficients (+L1 Regularisation for Term Pruning & Coefficient/Exponent Based Feature augmentation)

Tiny dependency-free calculator: fit an exponential decay to a trading signal's Information Coefficient and read off its half-life + rebalancing cadence. By microalphas.com

The old Fortran style C# coding repo for calculating probablities of superheavy nuclej alpha decay

PnL attribution and alpha decay framework for FinBERT sentiment signals — Information Coefficient decay curves, Spearman rank correlation, statistical significance bands, rolling signal quality monitoring.