Building Stock Factors and Filters in BigQuant AIStudio 3.0

Summary

This tutorial shows how to implement a collection of Chinese stock features and screening rules in BigQuant AIStudio 3.0. It divides them into expression features and expression filters, then explains that the same calculations can be entered as a SQL query. Features include rolling flow counts, return ratios, turnover averages, cash flow, moving-average ratios, SAR, volatility ratios, CCI, cross-sectional return ranks, and a Bollinger middle band. Filters cover profitability, price breakouts, money flows, sharp gains, and limit-up frequency.

The examples provide formulas and identify the platform tables and fields used, including how to estimate total cash flow from per-share cash flow and total shares. The article also suggests using the factors in an AI strategy. These are implementation examples rather than evidence of trading performance: it reports no backtest or returns, and warns that applying all the filters together may leave no eligible observations. The linked strategy is not described in detail here.

Key ideas

• AIStudio 3.0 supports implementing features and filters through either expressions or SQL.
• The examples combine price, volume, technical, financial, and money-flow data.
• Rolling calculations include moving averages, standard deviations, lagged returns, and event counts.
• Using every demonstrated filter at once may make the selection too restrictive.
• The document provides no performance evidence for the resulting factors or strategy.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.