Description

Have you ever wondered how visual attention influences stock price movements and investor behavior? In this enlightening episode of Papers With Backtest: An Algorithmic Trading Journey, we dive deep into the groundbreaking research paper titled "Acceleration Effect Combined with Momentum in Stocks" by Liwen Chen and Xinyi Yu. This study, which spans nearly five decades of data from January 1962 to December 2011 across major U.S. exchanges, uncovers the fascinating interplay between human psychology and market dynamics, revealing how investor overreactions can create profitable trading strategies.

The hosts dissect the innovative trading rules derived from this research, focusing on two pivotal strategies: the acceleration strategy and the deceleration strategy. The acceleration strategy capitalizes on stocks exhibiting rapid upward price trends, while the deceleration strategy takes a contrarian approach, betting against these trends. Our discussion highlights the significant backtesting results, demonstrating that the acceleration strategy not only outperformed traditional momentum strategies but also provided superior returns and enhanced risk-adjusted performance.

As we navigate through the complexities of visual patterns in trading decisions, we emphasize the robustness of these findings across various market conditions. The implications of visual attention in stock trading are profound, suggesting that recognizing price trends as they manifest in stock charts can unlock new avenues for enhanced trading opportunities. This episode is a treasure trove of insights for algorithmic traders, quantitative analysts, and anyone keen on improving their trading strategies.

Join us as we unravel the intricacies of visual attention, momentum, and the acceleration effect, equipping you with the knowledge to refine your trading approach. Whether you're an experienced trader or just starting your algorithmic trading journey, this episode of Papers With Backtest will provide you with valuable perspectives that could transform your understanding of market behavior and trading strategies. Don’t miss out on the chance to learn how to leverage psychological factors and visual cues in stock trading to enhance your performance!

Subscribe now and immerse yourself in the world of algorithmic trading, where data-driven insights meet practical application, and discover how the acceleration effect can reshape your trading landscape.

Hosted by Ausha. See ausha.co/privacy-policy for more information.

Transcription

• Speaker #0

  Hello. Welcome back to Papers with Backtest podcast. Today we dive into another algo trading research paper.

• Speaker #1

  And this one's quite interesting. It's called Acceleration Effect Combined with Momentum in Stocks by Liwen Chen and Xinyi Yu.

• Speaker #0

  Okay, let's unpack this. Have you ever looked at a stock chart that just seems to be taken off like a rocket?

• Speaker #1

  Yeah,

• Speaker #0

  absolutely. Well, the core idea here seems to be that those sort of visual patterns in historical stock prices really grab investors' attention.

• Speaker #1

  Right. And the authors argue this attention can lead to overreactions. And what's really interesting for us, potentially profitable trading strategies, maybe even better than standard momentum.

• Speaker #0

  And they didn't just look at a couple of years either. This study covers a massive period, January 1962, all the way to December 2011.

• Speaker #1

  Yeah, that's nearly 50 years of data across the big U.S. exchanges to NYSE, Amex and Nasdaq. It's a real goldmine.

• Speaker #0

  Exactly. So our mission today in this deep dive is to really get into the specific trading rules they came up with and, you know, the crucial part with the back test showed.

• Speaker #1

  Let's do it. So they start with this idea of visual attention. Why do these patterns matter? They link it to gestalt psychology.

• Speaker #0

  Gestalt. OK, refresh my memory. What's the connection to stock charts?

• Speaker #1

  Well, the gist is our brains love finding patterns. The law of continuity specifically suggests we expect trends to continue. So. you see a price accelerating upwards.

• Speaker #0

  You instinctively think it'll keep accelerating.

• Speaker #1

  Exactly. It's not just data points. It's a visual narrative of acceleration that people latch onto.

• Speaker #0

  Okay, that makes intuitive sense. But how did they actually measure this visual pattern? It wasn't just eyeballing charts,

• Speaker #1

  right? Oh, no, much more systematic. First, they do the standard momentum sort, rank stocks into five groups, quintiles based on past, say, J-month returns. Yeah. Your winners and losers.

• Speaker #0

  Standard stuff. Top 20% winners, bottom 20% losers. Got it. What's next?

• Speaker #1

  Here's the clever bit. For every single stock, they look at its daily prices over those past J months and run a regression.

• Speaker #0

  A regression.

• Speaker #1

  Yeah. They regress the price against time, tin, and times square, too. The equation is basically price alpha plus beta plus gamma. Ah,

• Speaker #0

  okay. And I'm guessing that gamma, that's the crucial part for acceleration.

• Speaker #1

  You got it. Gamma tells you about the curvature, the convexity. Positive gamma means the price curve is bending upwards faster and faster, accelerating. Like a car speeding up.

• Speaker #0

  And negative gamma.

• Speaker #1

  Decelerating. The rate of change is slowing down. Could be an increase slowing or a decrease slowing.

• Speaker #0

  So within the winners and losers, they sort again, but this time using this gamma value.

• Speaker #1

  Precisely. Into another five quintiles based on gamma. So you get winners accelerating the most, winners decelerating the most, losers accelerating downwards fastest, and losers whose fall is slowing down.

• Speaker #0

  That gives them a lot of combinations to play with. They design nine strategies based on this.

• Speaker #1

  Nine in total, yeah. Strategy one is just your plain vanilla momentum, buy winner, sell losers. But the others mix in this acceleration factor.

• Speaker #0

  Let's focus on the key ones that really highlight their idea. Strategy six, the acceleration strategy, and strategy seven, the deceleration strategy. How does strategy six work?

• Speaker #1

  Strategy six is pure acceleration focus. They buy the winners, showing the most acceleration top 20% gamma within the winners. And they short the losers, showing the most acceleration downwards bottom 20% gamma among the losers.

• Speaker #0

  So betting on those visually strong trends to keep going strong.

• Speaker #1

  Exactly. Riding that visual momentum, you could say.

• Speaker #0

  Okay. And strategy seven, the deceleration one, is that the opposite bet?

• Speaker #1

  Pretty much. It's betting against those accelerating trends continuing. They buy winners whose price gains are slowing down bottom 20% gamma in winners. And they sell losers whose price drops are also slowing top 20% gamma in losers.

• Speaker #0

  Kind of a contrarian play within momentum, fading those visually strong moves that seem to be losing steam.

• Speaker #1

  You could see it that way, yeah.

• Speaker #0

  All right. The moment of truth. The back tests. What happened when they ran these strategies? Let's use their main example. Rank on 12 months, hold for six months. J12, K6.

• Speaker #1

  The results were pretty striking. Plain momentum, strategy one, did okay. 83.22 basis points per month. That's roughly 10.5% a year. Not bad.

• Speaker #0

  Okay. The benchmark. How did the acceleration strategy stack up?

• Speaker #1

  Strategy six, the acceleration one, blew it away. 132.46 basis points per month.

• Speaker #0

  Wow. That's, let me calculate, over 17% annually.

• Speaker #1

  Yeah. A really significant jump. Almost 7% more per year than standard momentum in this test.

• Speaker #0

  And strategy seven, the deceleration one, betting against the curve.

• Speaker #1

  That one? Well, it had the lowest return, only 46.80 basis points per month. Suggests that fading those visually strong trends wasn't the winning play here.

• Speaker #0

  Interesting. So betting on acceleration paid off much better. But what about risk? Higher returns often mean higher volatility. How did the Sharpe ratios look?

• Speaker #1

  Good question. The acceleration strategy also won on risk-adjusted terms. Sharpe ratio of 0.260 for strategy 6 versus 0.204 for plain momentum.

• Speaker #0

  So better returns and better risk-adjusted returns. That's compelling.

• Speaker #1

  It definitely suggests the acceleration factor adds something meaningful beyond just higher raw returns.

• Speaker #0

  Now, is this just a lucky period or set up? Did they check if it holds up? Robustness checks?

• Speaker #1

  They did. quite a few. The findings were consistent across different subperiods within that long 1962 to 2011 time frame. They held up across different exchanges too. Okay. And they checked using midpoints of badass quotes, you know, to make sure it wasn't just bid-ask bounce noise. They also removed January returns of the January effect, didn't change the main conclusion. And results were robust to different ranking and holding periods as well.

• Speaker #0

  Sounds pretty solid then. Not just a fluke. The outline mentions mean difference tests. confirming the outperformance was real.

• Speaker #1

  Yeah. Table 3, Panel B, they formally tested the difference between Strategy 6 and Strategy 1 returns. And yes, the acceleration strategy's outperformance was statistically significant. OK.

• Speaker #0

  What if you looked back even further than 12 months to rank them, say 24 months? Did that change anything?

• Speaker #1

  Actually, Appendix 1 shows that using longer ranking periods like J24 often led to even higher returns for the acceleration strategy. Real? And get this. For some of those longer ranking periods, the sharp ratio for the acceleration strategy was almost double that of plain momentum.

• Speaker #0

  Double. Wow. So that visual effect might build up over longer trends. See,

• Speaker #1

  it's plausible, yeah. The effect appears stronger over longer lookbacks.

• Speaker #0

  OK, but is this acceleration strategy just finding the absolute highest momentum stocks anyway? Is it just like a fancier way to pick extreme winners?

• Speaker #1

  They checked that, too. Looked at the overlap between stocks picked by strategy six. And a simple extreme momentum strategy like top bottom 4%. Yeah. The overlap was surprisingly low, only about 20, 25%.

• Speaker #0

  Huh. So it's genuinely picking a different subset of stocks based on that gamma, that curvature.

• Speaker #1

  Exactly. It's not just relabeling extreme momentum. It's capturing something different related to the pattern of returns.

• Speaker #0

  Did they try to control for other known factors like size, value, volatility, maybe the 52 week high effect? Does acceleration still matter then?

• Speaker #1

  Yes. They ran FOMM and kept progressions, table four. Right. They threw in controls for size, book-to-market, volatility, nearness to the 52-week high, all the usual suspects.

• Speaker #0

  And the acceleration dummies.

• Speaker #1

  Still significant, both the dummies for accelerating winners and accelerating losers held up. Suggests this visual pattern has predictive power even after accounting for those other effects.

• Speaker #0

  That's quite powerful. It means the look of the trend matters independently. What about standard factor models like Fama French? Can they explain these extra profits?

• Speaker #1

  That's in Table 5. And the answer is no. The Fama-French three-factor model couldn't explain the alpha from the acceleration strategy. Neither could models adding momentum itself or the 52-week high factor.

• Speaker #0

  So the acceleration profits are distinct from those known factors.

• Speaker #1

  Not only distinct, but here's the kicker. They found evidence that the acceleration strategy could actually help explain some of the profits of plane momentum and the 52-week high strategy.

• Speaker #0

  Whoa, hang on. It explains them, not the other way around.

• Speaker #1

  That's what the results suggested. It points towards this visual acceleration potentially being a more fundamental driver underneath some of those other effects.

• Speaker #0

  That is a really interesting finding. Okay, what about longer term? Minimum eventually reverses, right? Did they look at that?

• Speaker #1

  They did in Table 6. Left out up to 60 months. And yes, reversals happen, typically starting around the 24-month mark.

• Speaker #0

  Consistent with other momentum research, anything different about the acceleration stocks?

• Speaker #1

  Yes. The reversals were larger, and they happened sooner, for the stocks identified by those acceleration strategies like 5, 6, and 8.

• Speaker #0

  So a bigger boom followed by a bigger bust, maybe.

• Speaker #1

  Kind of. It fits their narrative perfectly. The visual acceleration leads to a stronger initial overreaction, which then results in a more pronounced correction later on.

• Speaker #0

  Makes sense. Did they give any color on what types of companies these accelerating winners and losers tend to be?

• Speaker #1

  A little bit in table seven. Accelerating winners tended to be larger firms, less volatile, more liquid. OK.

• Speaker #0

  And the losers accelerating downwards.

• Speaker #1

  Those tended to be smaller. More volatile and less liquid, especially compared to losers whose fall was decelerating. Right.

• Speaker #0

  And just quickly, did it work consistently, like through recessions and expansions?

• Speaker #1

  Table 8 looked at that. Performance held up across different subperiods and business cycles. Profitability was generally higher in expansions, which isn't too surprising, but the relative advantage of the acceleration strategy seemed consistent.

• Speaker #0

  So not just an artifact of a particular economic climate?

• Speaker #1

  Doesn't seem like it. And they also double-checked using midpoint quotes to rule out bid ask bounce effects in table nine. The main results held there too.

• Speaker #0

  And finally, they had a few other checks mentioned on page 38, excluding certain stocks or tweaking the timing.

• Speaker #1

  Yeah, they found profits were often even higher if they didn't skip a month between ranking and holding. Also stronger if they excluded penny stocks or micro caps. And interestingly, the effect seemed particularly potent for Nasdaq stocks.

• Speaker #0

  OK, lots of digging there. So wrapping this all up, what's the big takeaway for you from this deep dive?

• Speaker #1

  For me, it's that the visual pattern, the acceleration of the price trend seems genuinely important. It's not just that a stock went up, but how it went up visually. This appears to capture investor attention, lead to overreaction.

• Speaker #0

  And create these measurable... potentially superior trading opportunity.

• Speaker #1

  Exactly. The acceleration strategy they outline looks like a more focused, maybe more powerful way to harness momentum than just the standard approach. It adds a new dimension based on how trends look.

• Speaker #0

  It's fascinating to think that something like a visual illusion, how a chart curves, could be a key driver behind a major market anomaly like momentum.

• Speaker #1

  It really is. And the authors make a strong case that this kind of visually driven overreaction provides a a pretty compelling explanation for why momentum works.

• Speaker #0

  Thank you for tuning in to Papers with Backtests 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.paperswithbacktests.com. Happy trading.