Checking Dragon-Tiger List Returns for Lookahead Bias

Summary

The document raises a data-timing question about China’s Dragon-Tiger List, a report of stocks with notable trading activity. It names fields for returns one, two, and five days after a stock appears on the list, and says the five-day return is unavailable on the listing date but becomes available after five days. A strategy that filters on that field therefore cannot generate a next-day signal using it in real time, even though a later backtest can show signals.

The post asks whether this is a future-function problem but provides no investigation or answer. Its example highlights the need to distinguish when an event occurred, when a derived field becomes observable, and which date a backtest assigns that value to. Determining whether the data causes lookahead bias requires checking the platform’s historical data behavior and whether each value is exposed only after its publication or calculation date. No implementation details, verification results, or safeguards are supplied, so the issue remains unresolved.

Key ideas

• The listed dataset includes returns measured one, two, and five days after a stock appears.
• The five-day return is described as unavailable on the listing date.
• A strategy using that field cannot produce a real-time next-day signal from it on the listing date.
• A later backtest may display signals once the five-day return is available.
• The post asks whether this timing creates lookahead bias but gives no conclusion.

Tags

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