Skip to content
All library documents

Building StockRanker Backtests for Hong Kong and US Equities

Article BigQuant
Read the original

Summary

The document explains how to adapt a StockRanker multi-factor stock-selection workflow to Hong Kong and US equities on BigQuant. The backtest uses BigTrader’s custom historical data input, which must include date, instrument, and daily open, high, low, close, and volume fields. The Hong Kong example extracts market data, supplies chosen multi-factor features and a prediction label, and otherwise follows the platform’s existing StockRanker template. It mentions filtering out low-priced shares and low-volume stocks.

The US workflow follows the same structure, with a different daily data table and a field rename so stock names appear in results. The described US coverage is limited to constituents of the Nasdaq 100, Dow Jones, and S&P 500 indices. The document points to example strategy source, but does not specify the factors, label construction, training setup, portfolio rules, or performance results. It is therefore an implementation outline, not evidence that the model predicts returns or performs well out of sample.

Key ideas

  • BigTrader backtests require custom historical bars with date, instrument, and standard daily price and volume fields.
  • The StockRanker workflow combines selected factors with a prediction label for training and prediction.
  • The Hong Kong example filters out low-priced and low-volume names.
  • The US example uses index constituent data and renames a company-name field for result display.
  • The document gives no model settings, validation results, or evidence of strategy performance.

Tags

EquitiesChina marketsUS marketsMachine learningBacktesting

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