statistical-validation

Multi-agent trading platform on an Alpaca paper account: a CLI places orders on request; scheduled routines manage trailing stops, alerts and daily records. Agents on Microsoft Agent Framework. Research uses walk-forward, out-of-sample and multiple-testing checks; no tested signal (MACD+RSI, TSMOM, GEX) has an edge yet. Not financial advice.

896 tests / 42 scripts validating Zetyra calculators (GSD, CUPED, Bayesian, SSR incl. Single-Arm, RAR, Master Protocol). Replicates 8 published trials (Salk, DAPA-HF, PACIFIC, MONALEESA-7, I-SPY 2, STAMPEDE, REMAP-CAP, NCT03377023) + Leyrat 2024.

Reproducible audit of how sampling budget biases apparent state counts from PCA, TICA, and VAE analyses of molecular simulation data.

Falsify: a backtesting engine for MetaTrader 5 data, built so it cannot flatter a strategy. 600 configurations screened over two campaigns, 530 counted as attempts, zero survivors after correction.

Biomedical knowledge-graph reliability screening: source code, frozen protocol, external-data evaluation and derived evidence. Research software; no journal publication claimed.

Teste A/B com Python para validação estatística de um sistema de recomendação em e-commerce.

Independent psychometric audit of a synthetic professional certification examination, focusing on measurement quality, form comparability, and decision validity.

Multi-agent research system for Nasdaq-100 futures: autonomous paper trading gated behind a statistical validation harness (bootstrap CI, permutation tests, walk-forward). Honest negative results included.

Maximum entropy null models for bipartite networks

用合成对照数据检验“临界慢化/早期预警”分析流程是否可靠，避免虚假显著结果。Validate critical slowing down / early-warning pipelines on synthetic controls before trusting them on real data.

QuantForge is a modular, event-driven backtesting framework built for quantitative research — combining realistic execution, reproducibility, multi-asset portfolio testing, walk-forward validation, Monte Carlo analysis, and White’s Reality Check

Self-observing decision model for gold (XAU/USD): structure perception, regime models, abstention, decision journal and reflection, validated with pre-registered walk-forward gates and random controls. 906 tests.

AI-native quantitative research & validation platform — reproducibility, statistical validation, production-grade financial data engineering