- Speaker #0
Hello, welcome back to Papers of Backtest podcast. Today, we dive into another algo trading research paper. This time we're going international with a strategy that not only seeks to profit, but also potentially offers some protection when the markets take a downturn.
- Speaker #1
What's fascinating here is the concept of pairs trading applied to international ETFs. It's like finding those, you know, dynamic duos in the global market, those ETFs that tend to move together historically. Then. when they drift apart, the strategy bets they'll come back together, kind of like magnets drawn to each other. OK,
- Speaker #0
that's a great analogy. But before we get into the specifics of this paper, let's break down the basic mechanics of pairs trading for those who might be new to the concept.
- Speaker #1
Absolutely. Imagine a seesaw with one side representing the price of one asset and the other representing the price of another asset. In pairs trading, we're looking for assets that usually move up and down together, keeping that seesaw balanced. But every now and then, one side might dip lower or rise higher than usual. creating an imbalance and that's our cue.
- Speaker #0
So we're essentially betting that the seesaw will rebalance itself, meaning the prices will converge back to their historical relationship.
- Speaker #1
Precisely. The strategy exploits those temporary deviations, buying the undervalued asset and simultaneously shorting the overvalued one. The key is identifying those pairs that have a strong historical correlation but are currently out of whack.
- Speaker #0
That makes sense. Now let's dive into the paper itself. It's called Pairs Trading on International ETFs. by Tamakos, Wang, and Skisas. And it focuses on applying the strategy to, you guessed it, international ETFs.
- Speaker #1
And what's particularly interesting is the paper's emphasis on the specific trading rules they employed. These rules are like the recipe for the strategy dictating when to enter and exit trades to maximize profits.
- Speaker #0
So what are the key ingredients of this trading recipe?
- Speaker #1
The paper outlines a two-phase approach. First, there's a formation period of 120 days. where they track a basket of 22 international ETFs, representing various regions and sectors. During this phase, they're looking for those strongly correlated pairs, you know, those assets that tend to move in sync.
- Speaker #0
So they're gathering the intel, figuring out which ETFs have that dynamic duo relationship.
- Speaker #1
Exactly. Then comes the trading period, which lasts for 20 days. Now, here's where things get a bit more technical. To determine when to trigger a trade, they use a measure called absolute deviation. It might sound complicated, But it essentially measures how far the current price difference between the two ETFs in a pair has strayed from their historical average.
- Speaker #0
So instead of waiting for a massive price gap, they're looking for even slight deviations from the norm.
- Speaker #1
Yes. They set a threshold at 0.5 times the historical standard deviation of the price difference. If the absolute deviation exceeds this threshold, it signals a potential trading opportunity.
- Speaker #0
It's like they're setting up tripwires. waiting for those prices to stray just far enough to trigger a trade. Makes sense. But why use absolute deviation instead of the more common sum of squares method?
- Speaker #1
That's a great question. Absolute deviation is more sensitive to smaller price movements, which could lead to more trading opportunities, especially in a market where those smaller inefficiencies can add up over time. It's like casting a wider net to catch more fish.
- Speaker #0
Okay, so we've got our correlated pairs, our trading period, and our trigger mechanism. What happens next? How do they actually execute the trades?
- Speaker #1
Once the absolute deviation threshold is breached, they buy the ETF, whose price has dipped relative to its partner, and simultaneously short the ETF whose price has risen. It's like betting on that seesaw to rebalance itself.
- Speaker #0
So they're going long on the underdog and short on the overachiever, hoping that gap will close.
- Speaker #1
Precisely. The paper's backtest results show that this strategy, with these specific rules, yielded some impressive returns. In fact, the portfolio using the top... five most correlated pairs achieved an annualized Sharpe ratio of 1.86.
- Speaker #0
Wow, 1.86. Now for our listeners who might not be familiar with the Sharpe ratio, it essentially measures risk-adjusted returns. Anything above 1 is generally considered pretty good. So 1.86 suggests this strategy was not only profitable, but also relatively consistent.
- Speaker #1
Right. And to add further context, the strategy consistently outperformed a simple buy and hold of the S&P 500 over the same period. which suggests it was able to generate alpha, meaning returns that exceed the market's overall performance.
- Speaker #0
So they weren't just riding the market's coattails. They were actually beating it consistently. That's definitely piquing my interest. But before we get too carried away with these impressive results, I'm curious to know how the strategy performed during different market conditions. Did they break down the results by bull and bear markets?
- Speaker #1
They did. And that's where things get really interesting. The strategy not only performed well overall. but it truly shined during those turbulent bear markets.
- Speaker #0
Now, hold on. That's pretty counterintuitive, isn't it? Most strategies struggle during bear markets. What made this one so resilient? I'm all ears.
- Speaker #1
Let's unpack that in the next segment. Remember that short component we talked about? Well, it turns out it played a crucial role in the strategy's outperformance during bear markets.
- Speaker #0
So when the market was going down, the strategy was actually making money on those short positions. That's pretty cool.
- Speaker #1
That's right. The paper even breaks down the returns separately for the long and short sides of the trades. And the short side consistently outperformed, especially during downturns.
- Speaker #0
That's a fascinating insight. It suggests that the strategy could potentially offer some downside protection during those periods. When the broader market is taking a hit, it's like having a built-in shock absorber for your portfolio.
- Speaker #1
Exactly. And this ties back to the idea of pairs trading being a market-neutral strategy. It's not just about profiting from rising markets. It's about exploiting price discrepancies regardless of the market's overall direction. So even when things are looking gloomy, this strategy can potentially find opportunities to generate returns.
- Speaker #0
OK, so we've established that this strategy has the potential to both profit and protect, which is pretty compelling. Yeah. But let's dive back into those specific trading rules for a moment. Did they mention anything about how long these trades typically lasted?
- Speaker #1
Yes, they did analyze the average trade duration, and it varied depending on the specific pair and market conditions. But here's a key takeaway. The average holding period for both long and short positions was less than the 20-day trading period.
- Speaker #0
So they weren't holding onto these trades for the entire 20 days. That suggests a fairly active trading approach.
- Speaker #1
Absolutely. It seems like they were taking advantage of those short-term price fluctuations, capitalizing on those moments when the seesaw started to tilt and then quickly rebalancing as the prices converged.
- Speaker #0
Okay, so we've got our correlated pairs, our entry and exit signals, and our average trade duration. Is there anything else we should know about these specific trading rules?
- Speaker #1
There is one more important detail. Remember how we talked about using absolute deviation to trigger trades? Well, the researchers also explored different threshold levels to see how it impacted the results.
- Speaker #0
Ah, so they were fine-tuning the sensitivity of their tripwires. Yeah. Makes sense. Did they find a sweet spot?
- Speaker #1
They did. While the 0.5 standard deviation threshold performed well overall, they found that a slightly lower threshold of 0.4 resulted in more frequent trades and a higher Sharpe ratio.
- Speaker #0
So by being a bit more aggressive with their entry points, they were able to squeeze out a bit more alpha. That's interesting. It suggests that there might be a trade-off between frequency and profitability.
- Speaker #1
Exactly. And that's where the art and science of algo trading come into play. It's about finding that optimal balance between identifying those profitable opportunities and minimizing trading costs and slippage.
- Speaker #0
Okay. So we've covered the trading rules in detail and we've seen some impressive backtest results. But before we wrap up, I'm curious to know if the researchers explored any nuances related to the specific ETFs they used. For example, did they compare the performance of large-cap versus small-cap ETFs or ETFs from developed versus emerging markets?
- Speaker #1
They did. And you might be wondering, should you focus on small-cap or emerging markets for this strategy? Well, while a blend of large and small-cap ETFs performed best overall, the portfolio composed solely of small-cap ETFs had a slightly higher Sharpe ratio.
- Speaker #0
So potentially... a bit more inefficiency and thus more opportunity for alpha in those smaller markets.
- Speaker #1
That's one interpretation. Small cap stocks are often less researched and followed by analysts, so there's a chance that mispricings might persist for longer periods, creating more opportunities for a strategy like pairs trading to capitalize on those discrepancies.
- Speaker #0
Okay, that makes sense. And what about the developed versus emerging markets split? I'd imagine there would be some notable differences there, given the varying levels of market maturity and volatility.
- Speaker #1
You're absolutely right. As you might expect, the emerging markets portfolio showed higher returns, but also higher volatility. This aligns with the general understanding that emerging markets offer greater growth potential, but also come with increased risk.
- Speaker #0
It's that classic risk-reward tradeoff. But I'm curious, did they find any evidence that pairs trading was more effective or less effective in one type of market over the other?
- Speaker #1
Interestingly, while returns were higher in emerging markets, the Sharpe ratio for the developed markets portfolio was actually slightly better. better once you factor in the added risk. This suggests that the strategy can be successful in both types of markets, but the specific risk-return profile might vary, depending on the underlying assets.
- Speaker #0
Okay, so there are nuances to consider based on your risk tolerance and investment goals. What I'm hearing is this. If you're comfortable with more volatility and seeking potentially higher returns, emerging markets might be worth exploring. But if you prefer a more balanced approach, developed markets could offer a smoother ride.
- Speaker #1
Precisely. It all comes down to aligning your investment strategy with your individual WISC profile and financial objectives.
- Speaker #0
This has been incredibly insightful. I'm curious, though, if they ever tried to dig into WHY. The strategy works beyond just the statistical analysis. Did they explore any fundamental economic factors that might be driving these price discrepancies?
- Speaker #1
They did. And that leads us to some even more fascinating findings. Let's explore those in our final segment. So they tried to connect their findings to real world economic factors, like using the FOMA French factors, which, you know, consider things like market risk, company size and value. They also looked at macroeconomic indicators and even individual company fundamentals.
- Speaker #0
So they went beyond just the numbers. Yeah. And tried to understand what's actually driving these price divergences.
- Speaker #1
Exactly. And while some factors like a book to market ratio and earnings per share showed some statistically significant relationships in certain cases, the overall explanatory power was limited.
- Speaker #0
So it's not as simple as saying, oh, this ETF is undervalued because its underlying companies have a high book to market ratio. There's something else at play here.
- Speaker #1
That's right. It suggests that pairs trading might be capturing price discrepancies that go beyond what traditional fundamental analysis can fully explain.
- Speaker #0
So are they saying there's this element of investor behavior, maybe even a bit of market psychology, that's influencing these short term price swings?
- Speaker #1
That's one interpretation. Remember, markets aren't always perfectly rational. Fear, greed, and even just plain old herd mentality can sometimes cause prices to move in ways that don't fully reflect underlying fundamentals.
- Speaker #0
Okay, that makes sense. And this is where the concept of market inefficiency comes in, right? Yeah. If everything was perfectly efficient and prices always reflected all available information, there wouldn't be these opportunities to profit from these discrepancies.
- Speaker #1
Precisely. And the fact that this strategy performed particularly well during the bear market of 2000-2002, a period of heightened uncertainty and volatility further supports this idea.
- Speaker #0
You mentioned earlier that the short side of the trades really shined during that bear market. Do you think that's connected to this idea of market inefficiency?
- Speaker #1
It's certainly possible. During times of market stress, investors might overreact to negative news, pushing prices down further than they should, creating opportunities for those short positions to really pay off.
- Speaker #0
So in a way, this strategy isn't just profiting from price discrepancies. It's also capitalizing on those moments when the market gets a little bit irrational.
- Speaker #1
That's an interesting way to put it. It highlights the importance of understanding both the quantitative and the psychological aspects of the market.
- Speaker #0
This has been an incredibly fascinating deep dive into pairs trading on international ETFs. We've learned about the specific trading rules they used, seen some impressive backtest results, and even explored the potential role of market psychology in the strategy's success. What key takeaways would you leave our listeners with?
- Speaker #1
First, this research suggests that pairs trading can be a viable strategy for international ETFs. By carefully selecting those dynamic duos and implementing clear trading rules, It's possible to generate consistent returns while also managing risk.
- Speaker #0
Right. It's not a magic bullet, but it's a tool that can be added to the algo trader's toolbox.
- Speaker #1
Second, remember that the short side of the trade can be just as important as the long side, especially during those volatile market periods.
- Speaker #0
It's a reminder that pairs trading is a market-neutral strategy, designed to profit from both upward and downward price movements.
- Speaker #1
And finally, don't underestimate the role of market psychology. Those seemingly irrational swings in price can create opportunities for those who understand how to identify and exploit them.
- Speaker #0
Fantastic. Thank you for taking us on this deep dive. It's been a truly enlightening exploration of this fascinating trading strategy.
- Speaker #1
My pleasure. Always happy to share these insights with fellow algo enthusiasts.
- 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.
