# Quant Ml Trading

> Complete quantitative and ML-powered trading toolkit — strategy validation, genetic optimization, decay monitoring, reinforcement learning, signal aggregation, data science pipelines, and statistical/quant foundations. Use this skill for: "generate a backtest report", "create tearsheet", "equity curve", "Monte Carlo simulation", "strategy report", "performance tearsheet", "backtest results", "drawdown analysis", "strategy statistics", "risk report", "publish backtest", "PDF report", "HTML report", "walk-forward", "out of sample validation", "rolling optimization", "WFO", "anchored walk forward", "parameter stability", "robustness test", "rolling backtest", "adaptive optimization", "parameter reoptimization", "stress test", "bootstrap simulation", "worst case scenario", "parameter sensitivity", "confidence interval", "ruin probability", "survival analysis", "tail risk", "heatmap", "parameter sweep", "optimization surface", "3D surface", "parameter landscape", "which parameters matter", "robust parameters", "se

- Skill: `mahmoud20138/quant-ml-trading` (Agent Skill)
- Install (CLI): `npx skillmds@latest add mahmoud20138/quant-ml-trading`
- Raw SKILL.md: https://api.skillmd.com/api/skills/mahmoud20138/quant-ml-trading/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Web & Frontend
- Author: mahmoud20138 (https://skillmd.com/u/mahmoud20138)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/mahmoud20138/quant-ml-trading

---

> **Skill:** Quant Ml Trading  |  **Domain:** trading  |  **Category:** quantitative  |  **Level:** expert
> **Tags:** `trading`, `quant`, `machine-learning`, `genetic`, `reinforcement-learning`, `signals`


# Quant & ML Trading — Complete Toolkit

## Overview
Five powerful tools for quantitative strategy work:

1. **Strategy Validation** — backtest tearsheets, walk-forward OOS, Monte Carlo, sensitivity, A/B testing
2. **Genetic Optimizer** — evolve strategy parameters using evolutionary algorithms
3. **Decay Monitor** — detect when a live strategy is losing its edge
4. **RL Trade Agent** — reinforcement learning agent (DQN) that learns to trade from market replay
5. **AI Signal Aggregator** — combine signals from all strategies using weighted voting + ML meta-model

## Reference Files

| File | Contents |
|------|----------|
| `references/strategy-validation.md` | Tearsheet, walk-forward WFO, Monte Carlo, sensitivity heatmaps, A/B testing |
| `references/genetic-optimizer.md` | Gene encoding, fitness functions, GA engine, anti-overfitting safeguards |
| `references/tensortrade-rl.md` | Strategy decay monitor (CUSUM, rolling Sharpe, live vs backtest) + RL trading environment + DQN agent |
| `references/ai-signal-aggregator.md` | Weighted voting, ML meta-learner, confidence calibration, trade decision engine |
| `references/quantitative-trading.md` | Quant workflow diagram, performance metrics table, pitfalls table, key formulas (Sharpe/Kelly/Z-score/IR/Expectancy) |
| `references/statistics-timeseries.md` | Descriptive stats, return distributions, stationarity (ADF/KPSS), ARIMA models, GARCH volatility, regression, Fama-French factors, cointegration |
| `references/ml-trading.md` | Supervised learning (XGBoost/LSTM/RF configs), feature engineering, regime detection (K-Means/HMM/GMM), RL, NLP sentiment, model validation |
| `references/backtesting-execution.md` | 10 backtesting rules, walk-forward, Monte Carlo, detailed performance metrics, 8 pitfalls, TWAP/VWAP/IS execution, factor strategies |
| `references/data-science-pipeline.md` | DataPipeline (OHLCV clean/validate/resample), FeatureEngine, StatisticalAnalysis, ModelFactory, TradingDataStore (persistent storage) |

## Quick Decision Guide

```
Strategy tearsheet / backtest report / Monte Carlo?
  → Load references/strategy-validation.md

Parameter optimization / genetic algorithm / evolve strategy?
  → Load references/genetic-optimizer.md

Strategy decay / live vs backtest divergence / RL agent?
  → Load references/tensortrade-rl.md

Combine signals / meta-model / AI signal aggregation?
  → Load references/ai-signal-aggregator.md

Quant workflow / performance metrics / key formulas?
  → Load references/quantitative-trading.md

Stats / time series / ARIMA / GARCH / regression / Fama-French?
  → Load references/statistics-timeseries.md

ML models / feature engineering / regime detection / NLP?
  → Load references/ml-trading.md

Backtesting rules / execution algos / factor strategies (detail)?
  → Load references/backtesting-execution.md

Data cleaning / feature pipeline / anomaly detection / persistent storage / ETL?
  → Load references/data-science-pipeline.md

Multiple topics?
  → Load all relevant reference files
```

## Anti-Overfitting Rules (ALL skills)
1. Always use walk-forward OOS — never optimize on all available data
2. Monte Carlo bootstrap required before any live deployment
3. GA/RL results MUST be validated on unseen temporal data
4. Meta-model must be re-trained monthly with rolling walk-forward
5. Live vs backtest Sharpe gap > 0.5 = pause the strategy immediately
---

---

## Related Skills

- [Ai Pattern Recognition](../ml-trading.md)
- [Backtesting Sim](../backtesting-sim.md)
- [Statistical Analysis](../statistics-timeseries.md)
- [Algorithmic Strategies](../backtesting-sim.md)