How the day of the week affects stock market anomalies

How the day of the week affects stock market anomalies

April 25, 2016 Research Insights
Print Friendly
(Last Updated On: January 18, 2017)

Day of the Week and the Cross-Section of Returns

Abstract:

This paper documents a new empirical fact. Long-short anomaly returns are strongly related to the day of the week. Anomalies for which the speculative leg is the short (long) leg experience the highest (lowest) strategy returns on Monday. The exact opposite pattern is observed on Fridays. The effects are large; Monday (Friday) alone accounts for over 100% of monthly returns for all anomalies examined for which the short (long) leg is the speculative leg. Consistent with a mispricing explanation, the pattern is fully driven by the speculative leg of the strategy. The observed patterns are consistent with the abundance of evidence in the psychology literature documenting that mood increases from Thursday to Friday and decreases on Monday.

Alpha Highlight:

Evidence from psychology suggests that “mood” can  influence decision-making. For example, Wright and Bower (1992) find that forecasts are influenced by mood — good moods generate more optimist forecasts than bad moods. Researchers have found that mood, or sentiment, can influence financial decision making. For example, a negative emotional state has been found to induce irrational financial market behavior (Read: The Financial Costs of Sadness), and unpleasant weather affects our mood and leads to slower market reactions (DeHaan, Madsen and Piotroski, 2015).

This paper looks at how mood might be related to weekly trading patterns. The author first cites the psychological literature, which finds a mood pattern across day of the week:

  • Mood increases from Thursday to Friday
  • Mood decreases on Monday.

Anyone who has ever had a job understands this relationship. Mondays are a general drag, whereas Thursdays and Friday are more exciting because the weekend is closer!

But can mood affect weekly stock returns?

To address this question, the author first plots mood across the week using data from Golder and Macy (2011).  The chart below highlights that positive feelings increase throughout the week and negative feelings decrease throughout the week. Clearly, people’s moods are affected by the day of the week.

mood pattern across day of the week
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.

This paper finds two simple examples that support the idea that mood might affect financial markets:

  • VIX shows an 2.16% average daily increase on Monday, and an -0.7% average daily decrease on Friday…thus, decreasing sentiment is associated with increases in VIX.
  • The average returns on one-year Treasuries are nearly 4 times higher on Monday than on Fridays: Poor sentiment induces “flight to safety.”

Birru hypothesizes that sentiment-driven stock anomalies should exhibit return variation across days of the week in accordance with moods.

To test this hypothesis, he examines 14 long/short anomaly variables that are likely to be affected by investors’ mood changes. The sentiment-based abnormal returns should be attribute to the speculative leg, which is assumed to be the short leg of the long/short portfolio (in all cases except illiquidity and size (the short leg is the large-cap leg, which is less “speculative”)). For example, when looking at “age,” where the long leg is old stocks and the short leg is young stocks, it is assumed by the author that younger stocks are more affected by sentiment than old stocks.

Here are the anomalies examined (red = short leg is speculative; blue = long leg is speculative):

measurement

 

The below figure compares the monthly 4-factor alphas for the long-short anomalies on Monday and Friday. We can see there’s a strong, predictable variation in the cross-section of returns across separate days of the week.

The results strongly support the “mood-driven” hypothesis:

  • All anomalies, except size and illiquidity, have high returns on Monday. –> short leg is the speculative leg
  • By contrast, size and illiquidity anomalies have high returns on Fridays. –> long leg is the speculative leg
Day of the Week and the Cross-Section of Returns_alpha
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.

This paper shows that this return pattern is robust to different sub-sample periods, and is not explained by other factors that could also explain this empirical pattern: 1) macroeconomic news releases, 2) firm-specific news releases, or 3) institutional trading.

Conclusion:

Mood swings across the week will affect the returns of speculative stocks. Specifically, L/S anomalous portfolios for which the speculative leg is the short (long) leg experience the highest (lowest) strategy returns on Monday. The exact opposite pattern is observed on Fridays.

One aspect I really can’t understand is why “value” is not considered a “sentiment” driven anomaly. I would expect that the short leg (the expensive growth stocks) would be related to mood and speculation more than the long leg — the author doesn’t find this. The relationship is fairly flat, with a upward move on Friday. Weird.

value sentiment
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.

This paper is also related to a new Journal of Finance article called, “Return Seasonalities.” The finance literature is really getting hot and heavy on seasonality effects that move beyond January. About time…

Sounds like a job for a systematic approach!

When investing in financial markets, always ask yourself if you are Trying Too Hard? Behavioral bias is everywhere…


Editor’s note:

The author reached out with some feedback regarding our comment on “why “value” is not considered a “sentiment” driven anomaly.”

From Prof. Birru:

There is an explanation for this finding – which I briefly discuss in my paper, and is discussed in more detail in “Investor Sentiment and the Cross-Section of Returns” by Baker and Wurgler (2006). They similarly find that the B/M anomaly does not exhibit returns that are correlated with sentiment. In other words, they do not find evidence that growth stocks are more speculative than value stocks. It is intuitive to think that growth stocks are speculative stocks, but Baker and Wurgler (2006) point out that value stocks are also speculative because value stocks (high B/M) include some firms that are distressed, and distressed firms are particularly speculative. They argue that because both growth stocks and value stocks are speculative, both groups of stocks will have a similar sensitivity to sentiment, and therefore the return difference between growth and value stocks will not be sensitive to sentiment. The lack of a relationship between the B/M anomaly and day of the week is consistent with their interpretation and with their empirical findings.

 


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

Wesley R. Gray, Ph.D.

After serving as a Captain in the United States Marine Corps, Dr. Gray earned a PhD, and worked as a finance professor at Drexel University. Dr. Gray’s interest in bridging the research gap between academia and industry led him to found Alpha Architect, an asset management that delivers affordable active exposures for tax-sensitive investors. Dr. Gray has published four books and a number of academic articles. Wes is a regular contributor to multiple industry outlets, to include the following: Wall Street Journal, Forbes, ETF.com, and the CFA Institute. Dr. Gray earned an MBA and a PhD in finance from the University of Chicago and graduated magna cum laude with a BS from The Wharton School of the University of Pennsylvania.


  • evo34

    When the premise of an article is based on data published in 2011, I’m not sure it’s all that applicable to today’s market..

  • That is a nice hypothesis and one that can be tested. I’ll ask the team at quantopian

  • Andrew Greene

    I have a regression question (not article related) I was hoping you could answer. If an investor just bought some combination of factors that are used in regression analysis, but used extreme timing such that he would dramatically outperform or underperform a factor benchmark, would that show up as “alpha” in a regression?

  • You can test it. Grab the top value portfolio and run a regression on the 3 factors. See what you’re alpha is…then apply a long-term trend following model on the top value portfolio and run a regression…see what you’re alpha is…

  • Andrew Greene

    Thanks for getting me to run a regression. Never realized how easy that was, then found your old video on youtube. I tested HML and UMD vs a time series that picks the better of the returns from those two factors and found that the perfect timing model does have alpha. Assuming I did this correctly this is an interesting result I was not expecting.

  • evo34

    My point, actually, was to ask why one would write an article in 2016 based on data that only went to 2011? Bluntly, it impacts your credibility as a market analyst.

  • I think a lot of the tests are through 2013, but that is still a bit stale. One conclusion is the author is hiding something. But I’ll offer an alternative: writing academic papers can take 2-3 years depending on what else a professor has going on. So I’m not sure this is a credibility issue as much as it is probably a bandwidth constraint issue.

  • evo34

    Who care? Either way, it’s not going to be predictive if you are using stale data, or sourcing others who are uisng stale data. I would love to see blog post discussing the performance of IVAL and QVAL.