quant-trading-cn

What This Skill Does

The quant-trading-cn skill serves as a high-performance quantitative trading assistant designed for algorithmic strategy development, backtesting, and production execution. Drawing from extensive Indian market experience (Nifty 50/Midcap), it provides a structured framework to transition from conceptual strategies to robust, low-latency trading robots. It automates critical workflows including signal generation, tick-size management, volatility modeling, and portfolio rebalancing, while specifically addressing the common technical pitfalls that often cause discrepancies between simulated backtesting results and real-market performance.

Installation

To integrate this skill into your OpenClaw environment, execute the following command in your terminal:

clawhub install openclaw/skills/skills/guohongbin-git/quant-trading-cn

Ensure that you have the necessary execution environment configured to support the shell scripts (wizard.sh, universe-fetch.sh, check-code.sh) provided in the repository, as these scripts facilitate the primary interactive workflows.

Use Cases

• Strategy Development: Quickly prototype and iterate on trading algorithms using the built-in wizard. Define your risk profile, asset universe, and technical indicator parameters.
• Performance Optimization: Use the tool to debug existing trading bots, specifically targeting common failures like incorrect tick-size rounding or improper VWAP calculation resets that lead to order rejections.
• Market Analysis: Automate the retrieval of real-time market indices and stock pools, allowing for rapid adaptation of strategies based on current market breadth and liquidity.
• Cross-Market Adaptation: Utilize the systematic framework to migrate strategies from Indian market patterns to A-share (China) markets by adjusting for local trading hours and instrument characteristics.

Example Prompts

1. "Launch the quant-trading-cn wizard to help me build a momentum-based day trading robot for the Nifty 50."
2. "Run the check-code script on my trading_bot.py file and identify any potential issues with my order execution logic or tick size handling."
3. "Fetch the latest Nifty 100 constituent list using the universe-fetch script so I can refine my liquidity-based stock selection filter."

Tips & Limitations

• Production Caution: Always run code in a paper trading environment first. The skill provides educational guidance and structural frameworks but cannot account for unpredictable black-swan events or slippage in high-frequency scenarios.
• Data Integrity: Ensure that your backtesting data is clean. The system relies on accurate T vs T-1 alignment; mismatched timestamps will invalidate your P&L analysis.
• Tick Size Priority: 90% of order failures in automated trading stem from incorrect tick-size rounding. Always utilize the provided rounding logic before submitting an order to your broker API.
• Performance: For large datasets, utilize the Parquet caching and Polars vectorization features to achieve the documented 25x speed improvements compared to standard pandas-based approaches.