A-Share Multi-Factor Quantitative Screening

Problem

For individual A-share investors, a quote page shows prices but not whether a trading rule would have worked. Manually scanning limit-up names, MA golden crosses, and RSI oversold signals is also easy to miss. OneQuant 4.0 puts quotes, k-lines, conditional screening, backtesting, simulated trading, and risk checks into one local Flask service, reducing tool switching.

How It Works

• Data access: use AkShare for free market data, or configure JQData / RQData for longer history, covering indices, stock quotes, k-lines, northbound flow, and financial metrics.
• Entry points: the Web UI is for manual analysis; REST API is for Python batch backtests; CLI and scheduled jobs are for after-close automation.
• Core capabilities: screen limit_up, ma_golden, and rsi_oversold; run MA/RSI backtests; inspect equity curves, performance metrics, and signals; simulate trades in a local account and export Excel.
• Risk and data: review position ratios, stop-loss/take-profit settings, and risk overview; trade records and strategy parameters are stored in local JSON files by default, and the broker is simulated.

Boundaries

It is best for strategy validation and learning: single-user local deployment, HTTP polling, and non-millisecond latency. Backtest results still need assumptions for commissions, slippage, and stamp duty. For high-frequency trading, futures/options, deep HK/US stock analysis, or complex factor models, use a dedicated system.