How to Measure Momentum?
Since we’ve released our new book, Quantitative Momentum, we’ve received a handful of basic questions related to momentum–specifically as it relates to stock selection.
At this point, the so-called “momentum effect” has occupied academic researchers for several decades. Researchers have found that, on average, stocks with strong recent performance relative to other stocks in the cross section of returns tend to outperform in the future (see Levy 1967 for an old example and JT 1993 for a newer version). The effect has been well-documented by numerous follow-on researchers and the theory of “why” momentum works has been extensively explored (although we still don’t completely understand why it works).
So if an investor wants to harness momentum and implement it in the real world, a common question arises:
What is the best way to measure momentum for stock picking purposes?
The academic research response is to focus on so-called, “12_2 momentum,” which measures the total return to a stock over the past twelve months, but ignores the previous month. (e.g., Ken French data)
But why use 12_2 momentum? Why shouldn’t we use the 3-month momentum, or the 6-month momentum? Why 12-months? And why drop the most recent month’s returns? Let’s take these questions one at a time.
As we show here, a few perturbations of how far back in time we should look have already been tested by academics. The chart below is from Jegadeesh and Titman in 1993. They show that the best performing strategy ranks stocks on their past 12-months returns, and holds for 3 months (forming overlapping portfolios).
Thus, 12-month momentum looks reasonable.
Why drop the most recent month?
Notice that we look at 12_2 momentum, not 12_1 momentum. But why skip the most recent month?
The reason why relates to the short-term reversal effect associated with momentum.
There is an academic finding that short-term momentum actually has a reversal affect, whereby the previous winners (measured over the past month) do poorly the next month, while the previous losers (measured over the past month) do well the next month. We document the findings here. Researchers often argue that this is due to microstructure issues. Thus, most academics ignore the previous month’s return (or week in the case of JT 1993) in the momentum calculation, and we also do this in order to eliminate this short-term reversal effect when implementing the strategy. It should be noted, however, that including the previous month’s returns has a marginal affect on the performance of momentum.
For more insight on 12_2 momentum, we invite you to explore a more important factor in momentum investing — rebalance frequency — as shown in the table above, and in our post here.
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)