Attention Prop Traders: The first half hour of trading predicts the last half hour…

Attention Prop Traders: The first half hour of trading predicts the last half hour…

August 21, 2014 Research Insights, Momentum Investing Research
Print Friendly
(Last Updated On: January 18, 2017)

Intraday Momentum: The First Half-Hour Return Predicts the Last Half-Hour Return

Abstract:

In this paper, using intra data from January 4, 1999 to December 31, 2012, we document an intraday momentum pattern that the first half-hour return on the market predicts the market return in the last half-hour. The predictability is both statistically and economically significant, and is stronger on high volatile days, recession days and some macroeconomic news release days. We interpret the trading behavior of daytraders and informed traders as the economic driving forces behind the intraday momentum.

Alpha Highlight:

Jegadeesh and Titman (1993) initially documented that past winners continued to be winners and past losers continued to be losers. Their “momentum” finding has been studied and confirmed in follow on research.

Why is this paper different?

Most papers examine momentum at a monthly frequency. This paper asks a simple question: Is there a simple intraday momentum strategy?

The authors find that the first half-hour of trading affects the last half-hour of trading.

How do they test this? First, they look for some relationship between the first half-hour of trading and the last half-hour of trading (there are 13 half-hours of trading). They run a predictive regression where they regress the returns for the last half-hour of trading (r_13) on the returns from the first half-hour of trading (r_1) and the returns from the twelfth half-hour of trading (r_12). The data is based off the SPDR S&P 500 ETF Trust (SPY) from 1/4/1999 until 12/31/2012.

The results are shown below:

2014-07-29 11_46_12-SSRN-id2440866.pdf - Adobe Acrobat Pro
The results are hypothetical results and are NOT an indicator of future results and do NOT represent returns that any investor actually attained. Indexes are unmanaged, do not reflect management or trading fees, and one cannot invest directly in an index. Additional information regarding the construction of these results is available upon request.

The table above shows a positive loading on r_1 (as well as r_1 and r_12) for all time periods, so the returns from the first half-hour positively and statistically predict returns during the last half-hour. Additionally, the regressions yield an R-squared value of 2% over the entire time period, and 4.3% during the financial crisis.

Is there a way to trade this finding?

The authors suggest going long (short) the last half-hour if the market return is positive (negative) during the first half-hour. A second strategy would be to go  long (short) the last half-hour if the market return is positive (negative) during the twelfth half-hour. A third strategy would be to go long (short) the last half-hour if the market return is positive (negative) during the first and twelfth half-hour, otherwise earn zero return. The results are shown below:

2014-07-29 12_00_01-SSRN-id2440866.pdf - Adobe Acrobat Pro
The results are hypothetical results and are NOT an indicator of future results and do NOT represent returns that any investor actually attained. Indexes are unmanaged, do not reflect management or trading fees, and one cannot invest directly in an index. Additional information regarding the construction of these results is available upon request.

The table above shows that a simple intraday momentum strategy earns an annual return of 6.34% when trading simply off the returns of the first half-hour of trading. Compared to a strategy of always going long during the last half hour (“Always Long”) and a simple Buy-and-Hold strategy. The simple intraday momentum strategy outperforms.

The authors give two potential explanations for this finding: trading behavior of daytraders (disposition effect), and strategic trading of informed traders.

What do you think? Time to become a day-trader?


Note: This site provides NO information on our value investing ETFs or our momentum investing ETFs. Please refer to this site.


Join thousands of other readers and subscribe to our blog.


Please remember that past performance is not an indicator of future results. Please read our full disclosures. The views and opinions expressed herein are those of the author and do not necessarily reflect the views of Alpha Architect, its affiliates or its employees. This material has been provided to you solely for information and educational purposes and does not constitute an offer or solicitation of an offer or any advice or recommendation to purchase any securities or other financial instruments and may not be construed as such. The factual information set forth herein has been obtained or derived from sources believed by the author and Alpha Architect to be reliable but it is not necessarily all-inclusive and is not guaranteed as to its accuracy and is not to be regarded as a representation or warranty, express or implied, as to the information’s accuracy or completeness, nor should the attached information serve as the basis of any investment decision. No part of this material may be reproduced in any form, or referred to in any other publication, without express written permission from Alpha Architect.


Definitions of common statistics used in our analysis are available here (towards the bottom)




About the Author

Jack Vogel, Ph.D.

Jack Vogel, Ph.D., conducts research in empirical asset pricing and behavioral finance, and is a co-author of DIY FINANCIAL ADVISOR: A Simple Solution to Build and Protect Your Wealth. His dissertation investigates how behavioral biases affect the value anomaly. His academic background includes experience as an instructor and research assistant at Drexel University in both the Finance and Mathematics departments, as well as a Finance instructor at Villanova University. Dr. Vogel is currently a Managing Member of Alpha Architect, LLC, an SEC-Registered Investment Advisor, where he heads the research department and serves as the Chief Financial Officer. He has a PhD in Finance and a MS in Mathematics from Drexel University, and graduated summa cum laude with a BS in Mathematics and Education from The University of Scranton.


  • RandomDoc

    I searched the whole paper and the word ‘cost’ does not show up even once! A day trading strategy paper without a robust treatment of costs (including the often ignored but critical cost of taxes) is pretty meaningless. The 6.34% CAGR, relatively low to begin with, will evaporate in no time once costs are included.

  • Marco

    Costs and slippage are inportant but it depends in great amount on the broker. I believe the finding to be interesting not only because on the simple results on its own but to the use of leverage. Maybe replicating the same strategy exploiting margins with futures will show a great improvement on results. I do the same with little overnight edges in my trading:

    http://nightlypatterns.wordpress.com

  • maquant

    Besides the costs, I don’t believe the results. I tested this with ES Futures data. The results (without any costs/slippage) for the “12th” half hour as predictor for the “13th” is around zero. The combined succes rate of almost 77% success rate of the combined strategy is obviously crap.

  • Marco

    What do you mean for 12th? I think the paper is about using the first half hour after the open to predict the last half hour before the close.. Am I right?
    http://nightlypatterns.wordpress.com

  • that is definitely true!

  • Jack Vogel, PhD

    While the paper mainly examines the effect that that 1st half hour of trading has on the last half hour (13th), it also examines the effect of the 12th half hour on the last half hour (13th). Trading simply off the first half hour provides the highest CAGR (6.97%). Alternatively, trading a combined signal from the 1st and 12th half hours has a higher “success rate” (76.88%) and a lower standard deviation, albeit with a lower CAGR (4.88%). I hope that helps!

  • agree.
    We did some tests today and found the same thing over this past year.
    We also tried ‘fancier’ versions of the strategy such as cross sectional momentum…ie. sort 500 stocks on momentum first half hour and/or 12th half hour and only buy top 30/50/etc. Not much there. My guess is the market maker/hft folks have widdled this down to the bone–assuming it was ever there

  • maquant

    Great, I was sure to have done my tests (ES/S&P Futures Data from ’97) in the right way, but anyway it is always good to have it confirmed. I found similar results with SPY tick data which I have only back to 2009 currently.

    I think that parts of the paper are simply wrong when I look at my results. The 12th half hour has no edge historically. And the combination with the 1st half hour is not contributing anything. Means the 76.88 % success rate is simply wrong when I look at my data.

    The first 5 “half hours” of the day have historically some small edges as well as the last 15 minutes of futures data (3 p.m. to 3:15 p.m.) of the day before. In addition if you look at futures it would be better to hold the position until 3:15 p.m. futures closing compared to 3 p.m. cash closing.

    In general -as you said- the effect is shrinking in the past couple of years and is to small to be exploited for normal traders.

  • Marco

    Great Jack… Could you help me find the half hourly historical data? I need it to add this findings with some additional filters to my own overnight trading at http://nightlypatterns.wordpress.com

  • Jack Vogel, PhD

    Marco, you can find the data through Bloomberg.

  • Marco
  • Hey Marco,
    You are correct. On BB you can only drag data down for most recent 140 days via Excel API. I recommend “warehousing” the data from this point forward so you have easy access. We have a policy against giving out certain databases (not for proprietary reasons, but legal). That said, we are happy to point you in the right direction to data sources. In this situation, I think the past 140 days can tell you the real story on this strategy…it hasn’t worked recently…
    Sorry we can’t help more.

  • NoblesseOblige

    There’s a similar paper out that also deals with intraday momentum in FX markets. The authors use the same methodology as Gao, Han, and Zhou and find predictability in the RUB/USD market. Predictability seems to be mostly concentrated in periods of financial stress.

    http://papers.ssrn.com/abstract=2694985