Agentic coding skills for backtesting trading strategies using VectorBT. Supports Indian, US, and Crypto markets with realistic transaction cost modeling, TA-Lib indicators, QuantStats tearsheets, and 12 ready-made strategy templates.

scikit-learn-compatible time-series cross-validation: purging, embargo, combinatorial purged CV, and deflated Sharpe ratios

An implementation of Giuseppe Paleologo's Rademacher Antiserum, designed to assess strategy performance consistency through Rademacher complexity and RAS-adjusted Sharpe Ratios. This code evaluates strategy robustness by applying Rademacher random vectors for anti-overfitting analysis.

Helix Treasury System

端到端、可复现的 A 股量化研究平台：因子挖掘 → 防过拟合 → Barra 风险模型 → AI Agent 智能挖掘 → 组合优化与归因 → 模拟交易 → 成果展示 Dashboard。全链路可审计、可复现。

Adversarial co-evolution orchestrator: an executor LLM improves an artifact, a deterministic scorer judges it (keep-if-better via git), a validator LLM advises — until quality peaks. Off-the-shelf agent CLIs, walk-forward scoring, live web dashboard. General-purpose, not just trading.

Crypto-currencies (Bitcoin) trading & Transfer Learning applications

QuantJourney Backtester

Systematic intraday opening-range breakout strategy on US equities — 10-year validated backtest with walk-forward optimization, statistical robustness suite, and live paper-trading on Alpaca.

End-to-end quantitative trading system for Taiwan stocks — DoubleEnsemble model, 300+ factors, Walk-Forward backtesting, React dashboard

一个面向 AI Agent / 量化实习的多智能体量化研究平台 / A resume-ready multi-agent quantitative research platform for AI Agent & Quant internships

A股量化Alpha研究: 多因子选股 + LightGBM非线性因子挖掘, Walk-forward验证, 自动交叉验证检查

This project implements a Walk-Forward Optimization (WFO) strategy using Tree-Parzen Estimator (TPE) for hyperparameter optimization. It includes modules for backtesting, configuration, optimization, and reporting.

AI-driven hybrid LightGBM–Ridge model for dynamic financial time-series allocation with Streamlit UI.

PPO + SAINTv2 transformer reinforcement learning trading bot for BTCUSD M1 on MetaTrader 5. Walk-forward training, 3-action policy, institutional stress-test, real-time GUI.

Portfolio backtest engine with montecarlo simulation, walk-forward, efficient frontier, FIRE, charts, performance optimization, max drowdown with Ai suggest!

Quantitative forex strategy research engine — 10 strategies, Optuna optimization, walk-forward + Monte-Carlo validation, honest overfit/ruin analysis. Pure Python.

Prove a trading edge survives real costs out-of-sample, or kill it. An honest edge-validation harness for CME micro futures & prop challenges.

Quant research: early-breakout momentum on SP500 — 9.4% CAGR vs SPY's 7.9% on 26-year survivorship-bias-free backtest (1,081 historical constituents). Automated Alpaca paper trading. 305 tests. Not financial advice.

Falsification-first quant research: two pre-registered studies of spot money-flow signals for BTC perps — base study falsified; the funding-divergence variant closed inconclusive-leaning-falsified, a fake winner caught by bootstrap, permutation, and persistence gates.