Aggregating Intraday Signals into Daily Equity Factors
Summary
This forum post asks how to test high-frequency equity factors built from minute bars. It cites examples from a China International Capital Corporation multi-factor series: the average squared correlation between the high and low prices of 50-minute bars, and return momentum among the 50 bars with the highest volume. The central issue is whether factor testing should retain multiple observations within each trading day or reduce each factor to one daily value.
The author has heard that lowering frequency can reduce computational demands and asks whether daily aggregation is appropriate, including whether a mean would be a suitable summary. The post does not provide an answer, compare aggregation methods, or report tests. It is useful as a statement of a factor-construction and sampling question, but it does not establish which method works; the right choice would depend on the factor definition and intended holding or evaluation horizon.
Key ideas
- Minute-bar factor calculations can produce multiple observations per trading day.
- The post asks whether those observations should be tested intraday or aggregated to daily frequency.
- It names price correlation and high-volume-bar momentum as example factors.
- It raises the daily mean as a possible aggregation method but offers no evidence or recommendation.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.