Skip to content
All library documents

Jump and Continuous Beta Factors from Five-Minute Equity Data

Article SuperMind

Summary

This research summary describes two equity factors, jump beta and continuous beta, constructed from five-minute market data over the prior year. It reports that both factors showed stock-selection power across sample universes, including after industry neutralization, and that their results improved on conventional beta selection. The reported RankIC half-life is about nine months, suggesting relatively slow signal decay. The summary also says performance among large-cap stocks resembled results for the broader market.

The factors are highly correlated with conventional beta, yet reportedly retain predictive information after orthogonalization against it. After adjustment for nine factor groups and conventional beta, the document reports negative RankIC values and ICIR values for both signals, with results remaining notable after neutralization. It also notes associations between jump beta and valuation, illiquidity, and speculation measures, while continuous beta is especially associated with illiquidity. These are reported findings rather than independently reviewable evidence: the linked paper is not included. The authors caution that beta factors have been less compelling in A-shares overall and invoke international BAB results as a reason for possible longer-term relevance.

Key ideas

  • Jump beta and continuous beta are estimated from five-minute data covering the prior year.
  • The report says both factors show stock-selection ability across sample universes and after industry neutralization.
  • Their reported RankIC half-life is about nine months.
  • Both signals correlate with conventional beta but are said to add information after orthogonalization.
  • The summary draws on A-shares and international factor comparisons, but the underlying paper is not provided here.

Tags

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