Skip to content

Market Intraday Momentum: summary and key takeaways

The paper tests whether the S&P 500 ETF's return from the prior close to the first half-hour of trading predicts the return in the final half-hour of the same day. It does, both in sample and out of sample. The effect is larger on volatile days, recession days and macro news days, and it appears in ten other ETFs.

Gao, L.; Han, Y.; Li, S.Z. et al. · 2014 · SSRNSummary updated 20 November 2026
Published2014Working paper
Asset focusEquitiesIntraday
TypeStrategyEquity Markets & Premia
Reading time5 minthis summary

Market Intraday Momentum at a glance

The question
Does the market's first half-hour return predict its last half-hour return within the same trading day?
Dataset & period
SPY (S&P 500 ETF) trade data from TAQ, February 1, 1993 through December 31, 2013, intraday. Days with fewer than 500 trades are dropped. Each day is split into 13 half-hour returns, the first measured from the prior close. Ten other ETFs are tested from their inception dates to 2013.
Method
The authors regress the last half-hour return on the first half-hour return, alone and together with the second-to-last half-hour return (the twelfth). Out-of-sample forecasts use recursive regressions and are scored against the historical average. Economic value is shown through a long or short timing rule on the last half-hour and through a mean-variance allocation with risk aversion set by the authors.

Key findings of Market Intraday Momentum

  • The first half-hour return predicts the last half-hour return with an in-sample R2 of 1.6%, positive slope, significant at the 1% level.
  • Out of sample, the first half-hour predictor gives an R2 of 1.4%, and adding the twelfth half-hour return lifts it to 2.0%.
  • Predictability rises with first half-hour volatility: R2 of 0.6% in the low-volatility group versus 3.3% in the high group. It is also stronger in recessions (R2 of 6.6% with both predictors) and on FOMC minutes release days (R2 of 11.0%).
  • A sign-based timing rule on the last half-hour has an annualized Sharpe ratio of 1.08 against 0.29 for buy-and-hold.
  • A mean-variance investor using the first half-hour return gains a certainty equivalent return of 6.02% per annum over the historical-average benchmark.
  • The first half-hour effect also shows up in ten other actively traded ETFs, covering other US indices, international equity, sectors and a bond fund.

Robustness, as reported in the paper

  • Recursive out-of-sample tests against the historical average benchmark give positive R2 for the first half-hour predictor (1.4%) and for both predictors together (2.0%).
  • After subtracting bid-ask costs (post-decimalization sample), the timing strategy average return falls to 4.46% per annum. From 2005 onward it is 6.52% after costs versus 7.96% before.
  • The pattern holds with and without the financial crisis, though it is much stronger during the crisis. The paper also reports it across volume, trade size, liquidity, institutional holding and month-end splits.
  • A multivariate regression on all half-hours finds only the first and twelfth significant at the 5% level.

Caveats and limits

  • The sample ends in 2013, because the authors say the TAQ data available to them stopped there, so recent behavior is not covered (inferred from the stated data limit).
  • Cost analysis uses quoted spreads only and ignores commissions, and covers only the period after decimalization.
  • The explanations (infrequent rebalancing and late-informed trading) are not tested directly. The authors say other explanations are likely and that institutional trade timestamps were unreliable.
  • The strategies trade the ETF at the last half-hour with idealized execution (inferred); market impact of size is not modeled.

Items marked (inferred) are our reading of the design, not statements by the authors.

Why it matters

The paper gives a documented intraday time-series momentum pattern in index ETFs, linking early-day moves to late-day flow. Practitioners who study execution timing, closing auctions or short-horizon signals can use it as a reference for when and why late-day trading may follow the morning direction.

Questions about this paper

What is market intraday momentum?

It is the positive link between the market's first half-hour return, measured from the previous close, and its last half-hour return on the same day.

Does the predictability survive out of sample?

Yes. Recursive out-of-sample R2 is 1.4% for the first half-hour predictor and 2.0% when the twelfth half-hour return is added.

What explanations does the paper offer?

Investors who rebalance infrequently and trade near the close, and late-informed investors who act on morning news late in the day.

Where can I read the original paper?

The original is at its publisher: https://doi.org/10.2139/ssrn.2440866. We link to the paper and do not host the PDF.

Follow new quant research

Download MacroFinance market data as CSV or Excel, or pull it by API, and test the ideas in these papers yourself. New summaries are added as we review them.

Create a free account
Summaries are written by MacroFinance from the published papers, in our own words, for information only. They are not investment advice and may omit detail. Papers are not hosted here: each page links to the original at its publisher.