- Speaker #0
Hello, welcome back to Papers with Backtest podcast. Today we dive into another algo trading research paper.
- Speaker #1
Yeah, looking forward to this one.
- Speaker #0
We're revisiting a really interesting idea in trading, the 52-week high effect. You might remember back in 2004, George and Wang pointed out something fascinating.
- Speaker #1
Right, that stocks trading near their highest price in the past year tended to, well, do better going forward than those way below their guys.
- Speaker #0
Exactly. And And their initial thought, their theory, It was about anchoring bias. Yeah. Like investors sort of get anchored to that recent high number.
- Speaker #1
Yeah. And maybe don't react fully to new good news, you know. But what's great is how researchers kept digging into this.
- Speaker #0
Which brings us to today's paper. It's from 2011 by Hong, Jordan, and Liu. Industry information and the 52-week high effect.
- Speaker #1
And this one really tries to get at the why. Is this effect just a risk thing? Or is it more about how investors actually behave?
- Speaker #0
And they had a twist, right? Yeah. Looking at whether it's driven by individual company stuff or maybe broader industry information.
- Speaker #1
Precisely. And they found something pretty surprising that maybe an industry-focused strategy could be even more profitable.
- Speaker #0
Okay, let's unpack this. So George and Huang looked at individual stocks hitting highs, but Hong, Jordan, and Liu kind of zoomed out.
- Speaker #1
They did. They looked at industries, and they used a really long time frame, 1963 all the way to 2009. That's a lot of market history.
- Speaker #0
Definitely gives it weight. So our mission today is to really understand the trading rules they tested for both individual stocks and these industry groups and crucially what the back tests showed.
- Speaker #1
Right. What actually worked and how well.
- Speaker #0
So let's start with that original individual strategy. The one George and Huang outlined, which this paper also uses as a sort of baseline. How did that work? What are the steps?
- Speaker #1
It's pretty straightforward, actually. At the end of every month, you calculate a ratio for each stock. They called it pre-log.
- Speaker #0
Pre-log.
- Speaker #1
Yeah. It's just the current price divided by its 52 week high. Simple as that.
- Speaker #0
Price relative to the last year's high. Got it.
- Speaker #1
Then you rank all the stocks based on that ratio. The top 30%, the ones closest to their high. That's your winner group.
- Speaker #0
And the bottom 30%, the ones furthest away.
- Speaker #1
Those are the losers.
- Speaker #0
Okay. Winners and losers based on individual stock highs. What's the actual trade?
- Speaker #1
You buy the winners go long and you short the losers. Bet against them.
- Speaker #0
And how are the portfolios weighted?
- Speaker #1
Equal weighted in this setup, so every stock gets the same amount of capital initially and you hold these positions for six months.
- Speaker #0
Six months, then rebalance. Okay. And the big question, did this individual strategy actually make money over that long 1963 to 2009 period?
- Speaker #1
Well, according to their backtest, yes. It generated an average monthly return of 0.43%.
- Speaker #0
0.43% a month. Not huge, but it adds up over decades, I suppose.
- Speaker #1
It can, absolutely. But there's always the counter argument, right? Maybe it's just risk.
- Speaker #0
You mean maybe stocks near their highs are just inherently riskier, higher beta or something?
- Speaker #1
Exactly. Maybe that return is just compensation for taking on more market risk. That's one possible explanation.
- Speaker #0
Right. Which makes the industry angle even more interesting. Did they find a stronger signal there? How did they build that industry strategy? Sounds a bit more involved.
- Speaker #1
It is a few more steps. First, they needed industries. They used two-digit SIC codes, which basically grouped stocks into... 20 different industries. OK,
- Speaker #0
20 broad industry groups. Then what?
- Speaker #1
Then for each industry, each month, they calculated an average pre-lag for all the stocks within it. But importantly, it was a value-weighted average.
- Speaker #0
Ah, so bigger companies within an industry had more influence on that industry's average pre-lag score.
- Speaker #1
That's the idea. So an industry score reflected whether its larger constituents, on average, were near their highs.
- Speaker #0
OK, so now we have industry scores, not just individual stock scores. How do they form the portfolios then?
- Speaker #1
Similar idea, but at the industry level. They rank the 20 industries by their average prelag. The top six industries were the winners.
- Speaker #0
And the bottom six.
- Speaker #1
The bottom six were the losers.
- Speaker #0
Right. So the winner portfolio wasn't just the top stocks. It was all the stocks in those top six winning industries.
- Speaker #1
Correct. And the loser portfolio was all the stocks in the bottom six industries.
- Speaker #0
And the trading mechanics. Still, long winners, short losers, equal weighted portfolios. Held for six months.
- Speaker #1
Yes, exactly the same structure in terms of the trade execution and holding period. Equal weighted across all the stocks selected.
- Speaker #0
And here's the kicker for you listening. What did the back test show for this industry approach? Was it better?
- Speaker #1
It was significantly better, actually. The industry strategy yielded an average monthly return of 0.60% over that same period.
- Speaker #0
0.60% compared to 0.43% for the individual one. That's What, nearly a 50% increase?
- Speaker #1
Roughly, yeah. A pretty substantial jump. And crucially, this higher return was statistically significant. It wasn't likely just luck.
- Speaker #0
Wow. Okay, so focusing on industry highs seems, at least initially, much more powerful. But back to that risk question. Did they check if this bigger return was just because the industry strategy was maybe, I don't know, loading up on more risk?
- Speaker #1
Absolutely critical question. And yes, they did extensive checks for risk. They use things like the Carhartt four factor model.
- Speaker #0
Right. That looks at market size, value and momentum factors.
- Speaker #1
Exactly. And they also use something called DGTW benchmark adjusted returns, which is another way to compare returns against similar stocks to account for characteristics.
- Speaker #0
So they really tried to isolate if there was genuine alpha. you know, excess return beyond just risk compensation. What happened when they applied these risk controls?
- Speaker #1
This is where it gets really interesting. For the individual stock strategy, its profitability pretty much vanished after these risk adjustments. Ah,
- Speaker #0
so the 0.43% monthly return, mostly explained by risk factors like size or momentum.
- Speaker #1
It seems that way. The DGTW adjuster returns weren't significant. And the four-factor alpha for the long-short portfolio wasn't statistically significant either.
- Speaker #0
But the industry strategy, did that hold up?
- Speaker #1
It did. Even after controlling for the Carhartt factors and using DGTW adjustments, the industry strategy still showed significant positive abnormal returns.
- Speaker #0
How much?
- Speaker #1
Let's see, 0.38% monthly using DGTW, still statistically significant. And the four-factor alpha was 0.22% per month, also significant.
- Speaker #0
So the industry effect seems much more robust to risk explanations. That's a key finding for you. And they mentioned the profit came mostly from the long side.
- Speaker #1
Yes, that's another important detail. The buy side buying the stocks and the winning industries seem to be driving most of that persistent alpha.
- Speaker #0
Interesting. So it's more about picking the right industries to buy than shorting the losers effectively. Did they look at other risk checks like mean adjusted returns?
- Speaker #1
They did. They compared stock returns to their own historical averages. Same story, basically. Individual strategy profit disappeared. Industry strategy profit remains significant. It really bolsters the idea that something beyond standard risk factors is at play with the industry effect.
- Speaker #0
OK, so if it's not just risk, maybe it is that anchoring bias idea George and Wang first proposed, but maybe it operates more strongly at the industry level. How did the paper investigate that?
- Speaker #1
They looked at what institutional investors are doing. The thinking is if there's underreaction or anchoring, Maybe sophisticated investors are noticing and trading on it.
- Speaker #0
Makes sense. What did they find?
- Speaker #1
They found that institutional investors, particularly the more active transient ones, did tend to increase their holdings in stocks and industries that were getting closer to their 52-week highs.
- Speaker #0
And decrease holdings when they fell away from the highs.
- Speaker #1
Exactly. Their behavior seemed to align with capitalizing on this effect, which lends support to the underreaction or anchoring explanation rather than just risk. It's like they see the market underreacting and step in.
- Speaker #0
That suggests it's a broader behavioral phenomenon, not just a few naive investors getting stuck.
- Speaker #1
OK, so we have this robust industry effect potentially linked to behavior. How does it stack up against other well-known strategies like momentum?
- Speaker #0
Good question. Momentum is obviously related, right? Buying past winners. They specifically compare the industry 52 week high strategy to both individual stock momentum and industry momentum strategies.
- Speaker #1
Did the 52 week high effect just. disappear once you accounted for momentum?
- Speaker #0
No, actually. The paper found that the industry 52-week high strategy remained profitable even when controlling for these momentum effects. It suggests it's capturing something distinct. So it's not just a different flavor of momentum.
- Speaker #1
It seems to be its own thing. They even ran simultaneous regressions, including both momentum and the 52-week high variables, and the industry high strategy stayed independently profitable.
- Speaker #0
That's pretty compelling evidence for it being a separate effect. What about long-term performance? Sometimes momentum strategies reverse over longer horizons. Did they see that here?
- Speaker #1
Interestingly, no. They looked at returns over longer periods, like three to five years out. While there's some evidence that individual momentum can reverse, they found no significant long-run reversal for the industry 52-week high strategy.
- Speaker #0
Which fits better with an underreaction story, right? If it were overreaction, you might expect it to correct itself eventually.
- Speaker #1
Exactly. Under reaction implies the market slowly catches up rather than then overshooting and then snapping back.
- Speaker #0
The paper also tried to pin down what kind of information drives this. Is it firm specific news or broader industry trends?
- Speaker #1
Yeah. They looked at whether the effect was stronger for stocks that tend to move more with their industry, those with high industry betas and high R-squared values relative to industry returns.
- Speaker #0
Yeah, was it?
- Speaker #1
Yes. The 52-week high effect, particularly the industry one, was more pronounced for these stocks whose prices seemed more driven by industry-level information rather than just company-specific news.
- Speaker #0
That really reinforces the industry information part of the paper's title. It seems the signal is strongest when reflecting broader industry sentiment or trends relative to past highs.
- Speaker #1
Precisely. Now, they also looked at where this effect might be strongest, considering things like how much information is readily available about a company.
- Speaker #0
Ah, like price informativeness, where maybe anchoring bias would be more likely if prices don't immediately reflect all news.
- Speaker #1
Exactly. They hypothesized the strategy might work better for firms where prices are, let's say, less efficient or informative. Think smaller firms, younger firms, maybe stocks with higher trading friction or lower analyst coverage.
- Speaker #0
And is that what they found?
- Speaker #1
Yes. The industry, 52-week high strategy, generated significantly higher profits among firms with these characteristics, small size, youth, high price impact, low analyst following, low institutional ownership. Places where information might travel slower or be harder to interpret.
- Speaker #0
Which again points towards that behavioral underreaction explanation. If... If prices were perfectly efficient everywhere, you wouldn't expect this.
- Speaker #1
Right. It suggests the strategy thrives where there is a bit more informational friction or potential for investors to anchor on past prices.
- Speaker #0
What about practical implementation? Holding for six months and rebalancing? Did they check if the results were sensitive to that specific setup? Like what if you just bought and held or used different weighting?
- Speaker #1
They did look at robustness to the rebalancing frequency and weighting. They found profitability persisted even with, say, a buy and hold approach after the initial sort, at least for a while. And while equal weighting was the main focus, they showed it also worked using value weighting, particularly within small stocks.
- Speaker #0
So it doesn't seem entirely dependent on that exact six-month rebalance. And they did other robustness checks, too, like different time periods.
- Speaker #1
Yeah, they sliced the data into subperiods and found the industry strategy consistently outperformed the individual one across different decades.
- Speaker #0
Even excluding major market events. Yeah. Like the dot-com bubble or the 2008 crisis.
- Speaker #1
Yes, they checked that too. Removing those extreme periods didn't eliminate the core finding. They also tested different holding periods like 3 months and 12 months, and the industry strategy still generally showed an an advantage.
- Speaker #0
It sounds like a pretty robust finding across different tests and timeframes.
- Speaker #1
It certainly seems that way based on their analysis over this long historical period. OK,
- Speaker #0
so let's distill the key takeaways for you, the listener. What does this research really tell us?
- Speaker #1
Well, first, this industry-focused 52-week high trading rule, which uses publicly available price data, showed surprisingly strong and persistent returns historically, significantly more so than just looking at individual stocks near their highs.
- Speaker #0
And second, the evidence seems to lean away from this just being about risk. It points more towards investor psychology specifically, anchoring or underreaction happening at the industry level.
- Speaker #1
Right. So for your own analysis, focusing on industry price action relative to the 52-week high might offer a more robust signal than the individual stock metric alone.
- Speaker #0
And finally, it seems the strategy works particularly well where information might be A bit fuzzier or less efficiently priced, like in smaller, less followed companies.
- Speaker #1
Exactly. Those seem to be the areas where this potential anchoring bias has the biggest impact on returns.
- Speaker #0
It's a fascinating alternative lens through which to view market behavior and potential opportunities.
- Speaker #1
Definitely provides food for thought.
- Speaker #0
Thank you for tuning in to Papers with Backtest podcast. We hope today's deep dive gave you useful insights. Join us next time as we break down more research.
- Speaker #1
And for more papers and backtests, find us at https.paperswithbacktest.com. Happy trading.
- Speaker #2
And just a final thought to leave you with. Could you potentially enhance this kind of strategy by layering in, say, industry level sentiment analysis or news flow alongside the pure price signal? And what are the practical hurdles for an individual trader trying to implement such a broad industry based approach? Something to ponder.
