Evaluating a Turnover Factor with Daily Rank Information Coefficients

Summary

The document demonstrates a simple cross-sectional factor evaluation for Chinese stocks. It queries daily records for 20-day average turnover and a five-day forward close-to-close return, then groups observations by date and calculates the Spearman rank correlation between the two measures. It summarizes the resulting daily information coefficients with their mean and an information ratio, defined in the example as the mean divided by the standard deviation.

This provides a basic way to inspect whether a liquidity-related characteristic ranks stocks in line with subsequent returns and how consistent that relationship is across dates. The example uses data from the first half of 2024, but reports no computed IC or IR values, so it establishes no factor result. It also gives no portfolio construction, transaction-cost, survivorship, or out-of-sample analysis. As written, the snippet may require correction: the replacement of infinities with missing values is not assigned back to the dataframe, and the displayed query contains a trailing comma before FROM.

Key ideas

• The example tests 20-day average turnover against five-day forward returns across stocks each day.
• It uses Spearman correlation to measure the daily cross-sectional rank relationship.
• The mean and standard deviation of daily information coefficients are used to calculate an information ratio.
• The sample query covers dates from January through early July 2024.
• No computed results or portfolio-level validation are provided, and the shown data-cleaning and query steps may need correction.

Tags

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