- Speaker #0
Hello, welcome back to Papers with Backtest podcast. Today we dive into another algo trading research paper.
- Speaker #1
And today we're tackling something called absolute strength momentum.
- Speaker #0
Right. It's an interesting concept. Basically, the idea is that stocks that have really moved a lot recently, up or down.
- Speaker #1
Big moves.
- Speaker #0
They tend to keep going in that same direction, at least for a little while.
- Speaker #1
Exactly. And it's important right off the bat to separate this from. Relative strength momentum, which I think more people are familiar with.
- Speaker #0
Yeah, relative strength is about comparing stock A to stock B or to the market.
- Speaker #1
Uh-huh. How's it doing compared to its peers? But absolute strength, that's just looking at the stock itself, its own history.
- Speaker #0
Has it gone up a lot on its own or down a lot?
- Speaker #1
Precisely. And, you know, it's not always obvious if the profits from, say, a standard relative momentum strategy are really because of that relative outperformance.
- Speaker #0
Or if it's actually these... big absolute moves doing the heavy listing underneath?
- Speaker #1
That's the question this research tries to unpack.
- Speaker #0
So that's what we're digging into, a specific trading rule based purely on this absolute strength idea.
- Speaker #1
It sounds simple on the surface. Buy the big winners, sell the big losers.
- Speaker #0
Right. Buy stocks that have shot up, short stocks that have cratered. Seems logical enough.
- Speaker #1
But the nuance is in defining big. What's a significant move?
- Speaker #0
And that's key here, isn't it? They don't just pick a number like, say, up 50 percent.
- Speaker #1
No, not arbitrary at all. They determine it endogenously based on a long history of actual return data.
- Speaker #0
OK, so they look back decades to see what a historically large move really looks like for stocks.
- Speaker #1
Exactly. And that's crucial because it avoids look ahead bias. You're not using future info to define past extremes.
- Speaker #0
Makes sense. Can't use tomorrow's data to trade today. So our mission here is to really get the rules down for this absolute strength strategy.
- Speaker #1
Understand how it works step by step.
- Speaker #0
And then critically look at the backtest results. Did it actually work?
- Speaker #1
Let's dive into those details. First, the data they used. It's pretty standard stuff for U.S. equity research. Common stocks from NYSE, Amex, and Nasdaq from the CRSP database.
- Speaker #0
And the time period.
- Speaker #1
The main analysis covers 1965 through 2016. So a good long stretch.
- Speaker #0
Over 50 years.
- Speaker #1
Yeah. But, and this is important for defining those big moves, They used data going all the way back to 1927 to build the historical context.
- Speaker #0
Wow. OK. Nearly a century of data for the baseline and the strategy itself. What are the mechanics?
- Speaker #1
They focused on what they call the 11-1-1 strategy.
- Speaker #0
11-1-1. OK. Break that down.
- Speaker #1
So each month you look back at a stock's return over the past year, but not quite the whole year. You look from month T12 up to month T2.
- Speaker #0
So the last 11 months, but skipping the most recent month, T1.
- Speaker #1
Exactly. That skip of the last month is pretty common in momentum research. Helps avoid short-term reversal effects.
- Speaker #0
Right. So you rank based on that 11-month return. What gets excluded?
- Speaker #1
Low-priced stocks, typically below a dollar. And stocks that didn't have enough return data during that 11-month window, they required at least eight observations.
- Speaker #0
Got it. Standard filters, and you form the portfolio.
- Speaker #1
Yes. Based on that ranking, you form your winner and loser portfolios, and you hold them for a month. Just one month.
- Speaker #0
And then rebalance and do it all again next month.
- Speaker #1
Rinse and repeat.
- Speaker #0
Okay, so that's the 11-month rank, one-month skip, one-month hold. Now, the crucial part, defining the absolute winners and losers from that rank, how they do that.
- Speaker #1
This is where that long 1927 onwards history comes in. They take a stock's 11-month return.
- Speaker #0
The T12 to T2 return.
- Speaker #1
Yes. And they compare it not just to other stocks right now, but to the entire historical distribution of all non-overlapping 11-month returns they have on record.
- Speaker #0
So comparing today's moves against like the entire history of 11 month moves.
- Speaker #1
Exactly. Think of it like a giant historical database of 11 month stock performance chunks. Where does the stock's recent performance fall in that grand scheme?
- Speaker #0
And winners are.
- Speaker #1
Winners are the stocks whose recent 11 month return puts them in the top 10% of that historical distribution. Truly exceptional performance compared to history.
- Speaker #0
And losers are the bottom 10% of that historical distribution.
- Speaker #1
Correct. Stocks that have performed exceptionally poorly relative to what's happened historically.
- Speaker #0
That feels much more grounded than just picking the top 10 percent of this month's performers, doesn't it?
- Speaker #1
It does. It anchors the definition of winner and loser to long term market behavior. And interestingly, they found these historical breakpoints were quite stable.
- Speaker #0
table How so?
- Speaker #1
Well, in the U.S. market, the cutoff for being an absolute winner over 11 months averaged around a 64% gain. Wow. And for losers, it was around a 43% loss. These numbers didn't wildly swing around.
- Speaker #0
Unlike relative strength, where the top 10% performer might actually have a negative return in a bad market, right? Precisely.
- Speaker #1
Or the bottom 10% relative performer might still be up in a roaring bull market. Absolute strength avoids that ambiguity.
- Speaker #0
Okay, so the definition seems robust. Now, the big question, did the strategy... buying these historical winners and selling historical losers. They call it Abs Mom.
- Speaker #1
Yes, Abs Mom.
- Speaker #0
Did Abs Mom actually make money at the back test?
- Speaker #1
It did, quite significantly, actually. Over that May 1965 to 2016 period, the average monthly risk-adjusted return was 2.50%.
- Speaker #0
Two and a half percent per month, risk-adjusted.
- Speaker #1
Yep. And the monthly Sharpe ratio was 0.34. Pretty solid numbers.
- Speaker #0
That's really strong. And was it consistent? Or just driven by a few good years.
- Speaker #1
Very consistent. They specifically highlighted the more recent period, 2000 to 2016.
- Speaker #0
Which includes the dot-com bust and the global financial crisis.
- Speaker #1
Exactly. Tough times for many strategies. But Abs Mom still pulled a 1.86% risk-adjusted return per month, with a sharp 0.20% even then.
- Speaker #0
That suggests real persistence. It wasn't just a fluke of the earlier decades.
- Speaker #1
Right. It weathered some major storms and still showed profitability based on this absolute momentum concept.
- Speaker #0
Now, they also mentioned something about... The Kvigorov-Smirnov test, KSD, what was that about?
- Speaker #1
Ah, yes, the KSD test. That was for a conditional version of the strategy. Basically, the KSD test measures how different the distribution of recent 11-month returns looks compared to the historical distribution.
- Speaker #0
So not just individual stocks, but the overall shape of recent returns versus history.
- Speaker #1
Exactly. Sometimes the whole market might be skewed towards extreme losses or extreme gains. The KSD flags those periods.
- Speaker #0
And what do they do when the KSD flags such a period?
- Speaker #1
When the KSD measure was very high in its top quintile, meaning recent returns looked very different from history, they suggested maybe switching the strategy off.
- Speaker #0
Switching it off, meaning?
- Speaker #1
Holding the risk-free asset instead. The idea is that in these highly unusual periods, the balance between winners and losers might be severely distorted.
- Speaker #0
Like you might have tons of losers and almost no winners. or vice versa.
- Speaker #1
Precisely, they give examples. End of February 2009, deep in the crisis. Far more absolute losers than winners. The market was just crushed. Looks so. Conversely, end of February 2004, lots more winners than losers. So the conditional strategy only ran the long-short absmom when the market's return distribution looked somewhat normal relative to history.
- Speaker #0
Did that conditional approach perform differently?
- Speaker #1
The paper suggests it helps manage risk, particularly around potential imbalances. But the core profitability comes from the Abs Mom signal itself. They also showed robustness by requiring a minimum number of stocks, like thirder, in both legs.
- Speaker #0
Right. So it wasn't driven by just a few outliers.
- Speaker #1
Correct. Even with that constraint, Abs Mom outperformed relative momentum.
- Speaker #0
And this wasn't just a U.S. stock phenomenon, was it? They tested it elsewhere. They did.
- Speaker #1
That's a crucial check. They looked at industry portfolios, corporate bonds, currencies, global equity indices.
- Speaker #0
Different asset classes.
- Speaker #1
Yes. And also international stock markets, UK, Europe, Japan, Canada.
- Speaker #0
And the results.
- Speaker #1
Consistently, absolute strength momentum seemed to perform better, more robustly than relative strength momentum across these different domains.
- Speaker #0
That's pretty powerful evidence for the concept itself. What about other robustness checks, different time periods, rules?
- Speaker #1
Oh, yeah, they really tried to break it. They looked at the pre-1965 data, 1927, 1964.
- Speaker #0
The earlier period.
- Speaker #1
Used different ways to calculate the historical breakpoints rolling windows, only using NYSE stocks for the breakpoints. Started the analysis later, like 1978.
- Speaker #0
Changed the timing.
- Speaker #1
Yep. Varied the ranking and holding periods, three months, six months, nine, twelve, instead of just the
- Speaker #0
11-1-1. And the conclusion through all that.
- Speaker #1
The absolute strength effect held up. It seems quite robust to these variations in methodology and time period.
- Speaker #0
That's convincing. Did they link it to anything fundamental about the companies?
- Speaker #1
They did touch on that. They found that the absolute strength winners tended to show, you know, significantly positive growth in things like ROE, ROA, earnings, gross profitability.
- Speaker #0
So real business momentum behind the stock momentum.
- Speaker #1
Kind of supports the idea, yeah. And the losers showed the opposite, significantly negative growth in those fundamentals. And importantly, these fundamental patterns were stronger for absolute momentum winner closers than for relative ones.
- Speaker #0
Any other quirks? Seasonality?
- Speaker #1
Briefly mentioned, yeah, some hints that institutions might track absolute strength, a possible November effect for absolute losers, maybe some window dressing pushing up big absolute winners in December. Standard seasonality patterns, but seen through this absolute lens.
- Speaker #0
And how does Abs Mom stack up against other maybe more complex momentum strategies we hear about?
- Speaker #1
Good question. They compared it to residual momentum, constant volatility momentum, dynamically weighted momentum. Some of the more advanced factors. Absmom held its own. Its Sharpe ratio was very competitive, often better than these more complex variations. Simple can be effective, it seems.
- Speaker #0
And crucially, what about momentum crashes? Relative momentum is known for those painful drawdowns.
- Speaker #1
That's a key finding. Absmom seemed to largely avoid the big crashes that played relative strength momentum, like the one in 2009.
- Speaker #0
So smoother performance profile.
- Speaker #1
Apparently so. Even when they excluded the specific months where relative momentum crashed badly, Abs Mom's strong performance remained. It suggests it's capturing a different, perhaps more fundamental aspect of momentum.
- Speaker #0
So to wrap up the findings then. This strategy, defining winners and losers based on their own historical performance extremes, seems to be significantly profitable.
- Speaker #1
Uh-huh. And robust.
- Speaker #0
Robust across time, across markets, across different ways of setting it up, and potentially less prone to crashing than standard relative momentum.
- Speaker #1
That's a pretty accurate summary of what the research indicates. It's a compelling alternative way to think about momentum.
- Speaker #0
Definitely gives you something to think about beyond just comparing stocks to each other. It's about the magnitude of the move itself.
- Speaker #1
Absolutely. Is the move historically significant on its own terms? That seems to be a powerful signal.
- Speaker #0
Thank you for tuning in to Papers with Backtest podcast. We hope today's episode gave you useful insights. Join us next time as we break down more research. And for more papers and backtests, find us at https.paperswithbacktest.com. Happy trading.
