purged-cross-validation

Leakage-aware financial ML pipeline using event-driven sampling, triple-barrier labeling, fractional differentiation, purged CV, XGBoost, meta-labeling, bet sizing, and backtesting.

Cost-aware 15m directional prediction for Binance BTCUSDT perp — pre-registered study, Cycle 1 = honest KILL-for-now (held-out sealed)

Crypto trading strategy backtesting & ML validation: triple-barrier labels, purged k-fold CV, deflated Sharpe ratio and a positive control. LightGBM, LSTM/BiLSTM, Binance + MEXC perpetual futures data.

Synthetic research-bias validation toolkit for systematic trading research.

Volatility forecasts, position sizing and purged walk-forward splits that never see the future. C11 core, numpy and scikit-learn front door.

Implementation of the Lopez de Prado AFML toolchain end to end: dollar bars, fractional differentiation, triple-barrier and meta-labelling, purged and combinatorial-purged cross-validation, deflated Sharpe, wired to an OpenBB data layer and walk-forward backtester. No strategy result is claimed.

코스피·코스닥 알파 심사 프레임워크 — 개별 트레이드 분포로 판정하고 랜덤 음성대조·purged CV·Deflated Sharpe 를 CI 가드레일로 강제. 기각 판정문까지 공개한다 · Alpha validation framework for KOSPI/KOSDAQ

15m crypto perpetuals edge research: triple-barrier + purged CV + deflated Sharpe harness with a positive control. Result: no edge after fees.