Mixing Momentum and Value: A Winning Combination?

Mixing Momentum and Value: A Winning Combination?

September 23, 2014 Value Investing Research, Momentum Investing Research
Print Friendly
(Last Updated On: January 18, 2017)

Combining Value and Momentum

Abstract:

This paper considers several popular portfolio implementation techniques that maximize exposure to value and/or momentum stocks while taking into account transaction costs. Our analysis of long-only strategies illustrates how a strategy that simultaneously incorporates both value and momentum outperforms a strategy that combines pure-play value and momentum portfolios that are formed independently. There are two advantages of the simultaneous strategy. The first is the reduction in transaction costs; the second is better utilization of unfavorable value and momentum information in a long-only portfolio. Our analysis also addresses the optimal way to combine the faster-moving momentum signal with the slower-moving value signal.

Alpha Highlight:

Numerous academic papers examine pure-play value and momentum portfolios. Sheridan Titman–one of the authors on this paper–has a well-established and respected group of papers on momentum. Why he decided to join in on this paper is a bit puzzling, given the paper’s weak robustness analysis and “practitioner-focused” bent. Nonetheless, the paper asks an interesting question: “Why not combine value and momentum in one portfolio and “kill two birds with one stone?” This paper is not the first to consider this issue. Asness, Moskowitz and Pedersen (2012) shows that even 50/50 equal combination of value and momentum can significantly outperform pure-play strategies. Asness (1997) and Daniel & Titman (1999) show that momentum and value are negatively correlated. Stivers and Sun (2009) use Return Dispersion (RD) as a proxy for volatility and thus generate a timing signal to combine these two strategies dynamically. The point of all this research is clear: momentum is cool; value is cool; combining the 2 is awesome. The authors of this paper use two simple methods to combine value and momentum and find the following benefits relative to “pure-play” value or momentum strategies:

  • Reductions in transaction costs
  • Higher Sharpe Ratios

The authors highlight that, “Momentum is a relatively fast moving characteristic, since returns from year to year are relatively independent, while value is a relatively slow-moving characteristic, since it is based on levels rather than changes in market values.” Thus, combining the fast and slow characteristics can better utilize the unfavorable information embedded in the characteristics. The two simple combination strategies are:

  1. Strategy 1 — Average V/M strategy: Ranks firms by Momentum and Value separately, and then compute the average rank (Rj), to calculate the stock’s Average V/M score.
    • The Avg. V/M score has exposure to both signals equally. The disadvantage of this simple strategy is that one signal has the ability to out-weigh the other. Thus, the changes in momentum scores have a large influence on trading, which lead to higher trading costs.
  2. Strategy 2 — Value|m>X: Rather than selecting stocks based on a combined signal, after initial positions are chosen, trades are initiated only when both value and momentum signals are sufficiently favorable.
    • This strategy has greater value exposure and less momentum exposure.

Which of the two combination strategies works better? Below is a main result table of the paper.  The table shows the performance of 5 different strategies from 2000 to 2013 (we focus on 4): Value only, Momentum only, Average V/M strategy, and Value|m>50% strategy.

2014-08-09 17_43_21-Combining Value and Momentum .pdf - Adobe Reader
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.

Key Findings: 

  1. From the table above, both the two combination strategies work well. They generate higher returns, lower standard deviations, and higher Sharpe Ratios than the Value or Momentum strategy only.
  2. The Avg. V/M strategy performs relatively better (highest Sharpe ratios for both large and small capitalization stocks) in this paper.

These two methods are easy to design and apply. But the data period in this paper is short, thus making the results not very informative or unique.


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 disclaimer. 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.


  • Value and momentum works quite well.

    See our back test study on European markets. We have combined earningsyield and 6 months momentum;
    Results can be found: on link below →

    http://www.quant-investing.com/strategies

  • Doug

    Looks like a lot of people are trying to find stocks that are good at everything – using value and momentum measures to build a scoring model al la Greenblatt’s work with value and quality (and your work as well). Wouldn’t it make more sense to build a pure value portfolio and a pure momentum portfolio, then combine them? (double-weighting the stocks that hit both portfolios).
    Put another way, using value/mom as screening factors can produce stocks that score, say, a 7 out of 10 for each score, with a total of 14 out of 20. Why not find a bunch of “10” value stocks and “10” momentum stocks?
    EDIT: I think Asness approached it my way. I’ll have to go back and check.

  • I’m personally a fan of building out best of breed value and best of breed momentum and then leaving it up to the investor to decide how/why they want x and 1-x exposure to value and momentum, respectively. The combo stuff muddies the process…but that’s my 2 cents

  • A value stock that starts to go up: Is that a value stock or a momentum stock ?
    → It is difficult convincing a value investor to use momentum factors…

    But nothing is just black or white…

    There are several evidence that momentum can be combined with value.

    It allows to avoid value traps, I guess.

  • A momentum stock if it has gone up a lot relative to all other stocks in the universe… and potentially a value stock if fundamentals have also gone up by a large amount. So you’re paying a much higher price, but on a fundamentals-to-price basis you haven’t moved much. That is just an example…

  • Steve

    You miss the point of the paper.
    Firstly, Asness is usually speaking in terms of long/short. Not always, but usually.

    Besides that; that’s what this paper tests…whether a combined approach is ‘better’ (using realistic trading assumptions) than a ‘split’ approach.

    I was personally amazed (as a fan of the split approach) that the combined approach does in fact, work better. The reason simply has to be (and is talked about in the paper) that the information that the other factor provides, is of value.

    Nonetheless, I’m still (with you, and Wes) in the split camp, as there are other aspects to consider also.

    For one; the authors don’t look at tracking error measures etc.

  • Chris Scott

    Interesting paper, but I’m not sure you can really draw any conclusions from it given the way the portfolios are formed and the lack of tests for statistical significance. Given that Asness (1997) finds that adding value to momentum provides no return differentiation, but adding momentum to value does provide return differentiation; it is plausible that there could be some way to combine value and momentum that would provide superior returns. I just don’t see an answer out of this paper. One interesting corollary question would be whether momentum adds anything to a value/quality portfolio like EQV. I’m not sure it would as quality may be identifying the characteristics that show up within value/momentum stocks.

    Edit:
    The more I think about this paper, the more I think there is significant data mining involved. The unusual method of portfolio formation and the limited number of breakpoints that are presented all point to optimization.

    As to value being a slow moving characteristic which provides a benefit when combined with fast moving momentum – this doesn’t make any sense. The slow moving characteristic of value is partial due the presence of forever marginal firms within the value portfolio (particularly when using B/P). Consider the company that is in decline, with declining revenue and negative earnings. It is cheap for a reason, has very poor prospects, and may eventually go bankrupt. It will stay in the value portfolio year after year until bankruptcy, never contributing a positive return. Value firms whose prospects improve have an increase in valuation and leave the value portfolio. How could this characteristic improve momentum returns? Evidence says it won’t. Momentum will improve the value return by eliminating some of these forever marginal firms, but value won’t help momentum. In the end isn’t it better to use quality to filter out the forever marginal firms from the value portfolio and incorporate momentum as a separate strategy?

  • Denys Glushkov

    As a matter of fact, momentum is not the best signal (at least according to backtests) to be combined with value. If you were to use a simple P/B as your value signal, then a simple equal-weighted combination (at the holdings, not return level) of P/B and a simple Free Cash Flow/Assets (updated quarterly) tends to nearly triple the IRs of a plain vanilla P/B strategy. The FCF/Assets combined with value also delivers lower turnover (nearly, half of that of Value/Momentum blend), higher ICs, albeit at the expense of relatively higher drawdown for LS portfolio. But FCF/Assets and Value is one of the most powerful signal combinations that i backtested.

  • Ronnie Shah

    “These two methods are easy to design and apply. But the data period in this paper is short, thus making the results not very informative or unique.”

    – Our paper goes back to 1975 [Just not Table 4 that was highlighted above] The point is not to evaluate what is the best factor to mix with Value, but rather how to think about a framework to mix fast and slow signals. We use value and momentum as two generic quant. signals to illustrate the various trade-offs. The actual paper has more breakpoints than the exerpt from Table 4 – we actually had more variations in breakpoints but the referee made us take it out.

  • Thanks for sharing the additional insights. Feel free to post the materials the referee made you take out. That will add some color/detail that readers would probably appreciate!
    Nice paper, BTW