Using Daily Popularity-Ranked Stock Lists in Backtests

Summary

This discussion addresses how to use a daily stock popularity ranking obtained through a query interface in a backtest when the research environment returns a dataframe but the backtest environment does not behave the same way. One suggested workflow is to retrieve each date’s ranking in the research environment, save the results, and load the corresponding file during backtesting. An alternative is to call the query during the backtest, using the last available trading date to form the request.

The discussion explicitly notes that the query interface has call limits and recommends precomputing and saving the daily data. It is a practical data-pipeline tip rather than a trading signal: the ranking is not evaluated as a predictor, and no returns, coverage checks, or historical availability details are supplied. A researcher using such popularity data would still need to verify date alignment, data availability at the simulated decision time, and any survivorship or revision effects before interpreting backtest results.

Key ideas

• Popularity rankings can be queried by date and incorporated into a historical backtest.
• One approach is to collect each day’s ranking in the research environment and save it for later use.
• Another approach is to query the ranking directly during the backtest using the last available date.
• The discussion recommends preloading data because the query interface has usage limits.
• The post does not evaluate the ranking’s predictive value or discuss historical data revisions.

Tags

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