Chinese Stock Ranking by Size and Dividend Yield with Three-Day Rebalancing

Summary

This BigQuant example builds a long-only Chinese stock portfolio from a filtered universe of main-board shares. It excludes suspended stocks and, in its feature filters, keeps non-ST stocks listed for more than a year, with positive trailing earnings and dividend yield, a float market capitalization cap, and a price-to-sales ceiling. A composite score weights cross-sectional ranks of total market capitalization, float market capitalization, and dividend yield, with dividend yield ranked in descending order.

The workflow extracts historical signals, selects ten holdings, assigns equal weights, and rebalances every three trading days. The backtest configuration uses daily bars, open prices for orders, a stated starting capital, and the CSI 300 as benchmark; the trading logic sells names outside the target list and adjusts positions for current targets. The document provides implementation settings but no performance results, so it does not establish that the ranking is profitable. Its size ranks may favor larger firms within the filtered universe, and the stated filters, transaction costs, data timing, and rebalance frequency all affect results.

Key ideas

• The universe is restricted to listed Chinese main-board stocks, with suspended shares excluded.
• The score combines ranks of total market capitalization, float market capitalization, and dividend yield.
• The filters require positive earnings and dividend yield, more than one year since listing, and specified valuation and size limits.
• The portfolio holds ten names at equal weights and rebalances every three trading days.
• The document gives backtest configuration but reports no performance evidence.

Tags

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