Transcription

• Speaker #0

  Hello, welcome back to Papers with Backtest podcast. Today we dive into another Algo trading research paper called Opposing Seasonalities in Treasury vs. Equity Returns. Now, you know we love digging into market anomalies here, and this one's a real head scratcher.

• Speaker #1

  It really is. This paper basically says that U.S. Treasury bonds, those safe haven assets we all know and love, actually have this crazy annual cycle in their returns. We're talking swings of over... 80 basis points between the peak and trough each year.

• Speaker #0

  80 basis points. That's huge. But here's where it gets even weirder. This cycle is the complete opposite of what we see in equity returns. So when stocks are zigging, bonds are zagging and vice versa, why would that be?

• Speaker #1

  Well, on the surface, it doesn't make a whole lot of sense. I mean, usually treasuries and equities move kind of in sync, right? So the researchers dug into this and what they found is super interesting.

• Speaker #0

  Okay, spill the beans. What's the secret sauce behind these opposing seasonalities?

• Speaker #1

  Get this. They're pointing to something called Seasonal Affective Disorder, or SAD. You know, that winter blues thing some people get when the days get shorter? The researchers think this might actually be playing a role in how investors behave.

• Speaker #0

  Hold on, are you saying that people's moods are messing with the bond market?

• Speaker #1

  It might sound crazy, but hear me out. Think about it. If you're feeling down and anxious during those dark winter months, maybe you're less likely to take big risks in the market. You might want to park your money in something safe and stable like treasuries.

• Speaker #0

  OK, I see where you're going with this. So as more people experience those sad symptoms in the fall and winter, they flock to treasuries driving up the prices.

• Speaker #1

  Exactly. And when demand goes up, returns tend to go down, at least in the short term. That's why we see that dip in Treasury returns during those months. Then when spring rolls around and the sunshine comes back, everyone's feeling more optimistic and risk tolerant. They start ditching those bonds and jumping back into equities, seeking those juicy returns.

• Speaker #0

  That's fascinating. But is there actually any evidence to support this SAD theory? Or is it just a bunch of speculation?

• Speaker #1

  The researchers actually put this to the test. They came up with a way to measure how many people are likely experiencing SAD symptoms at any given time. Think of it like a SAD thermometer for the market. They call this variable OROT.

• Speaker #0

  OROT. Got it. And what did they find when they looked at this SAD thermometer alongside market data?

• Speaker #1

  Well, they ran a bunch of regressions and found something pretty remarkable. This OMOT variable was statistically significant in both equity and bond regressions. In other words, it seems like there's a real link between the prevalence of SAD symptoms and those opposing market movements.

• Speaker #0

  So it's not just a coincidence. There's actually something statistically measurable going on here. Okay, I'm officially intrigued. But let's get to the part our algo minded listeners really care about. How can we turn this knowledge into a profitable trading strategy?

• Speaker #1

  Well, the researchers did some backtesting to see how a simple trading rule based on this seasonal pattern would perform. It's actually a pretty straightforward strategy. You simply buy treasuries in October, hold them until April, then sell and switch back into equities.

• Speaker #0

  OK, so you're basically riding that wave of SAD driven risk aversion, buying treasuries when they're in high demand and selling them when people are feeling more bullish. Sounds simple enough. But did it actually work?

• Speaker #1

  You bet it did. They ran this strategy on decades of historical data and found that it consistently generated average annualized excess returns of over 3%.

• Speaker #0

  3%. That's not bad at all, especially for a strategy that only involves trading twice a year.

• Speaker #1

  Exactly. And keep in mind, this is a very basic long only strategy. No shorting, no leverage, nothing fancy. The researchers even point out that with some further optimization, you could potentially boost those returns even higher.

• Speaker #0

  OK, this is where it gets really exciting for anyone interested in algo trading. So we've got this intriguing market anomaly linked to SAD and a simple trading rule that seems to exploit it quite effectively. But before our listeners rush out and start buying up all the treasuries they can find, let's dig a little deeper into the research and see if there are any caveats or nuances we need to consider.

• Speaker #1

  Now, before we get ahead of ourselves, it's important to understand why the researchers are so confident about this SAD link. Right.

• Speaker #0

  We can't just take their word for it. They have to have some pretty solid evidence to back up this claim.

• Speaker #1

  And they do. These researchers were super thorough. They didn't just jump to the SAD conclusion. They tested a ton of other potential explanations first, you know, just to make sure they weren't missing something obvious.

• Speaker #0

  So they were trying to rule out any other factors that might explain this weird seasonal pattern.

• Speaker #1

  Exactly. They started by looking at all the usual suspects like economic data, think changes in industrial production. unemployment rates, inflation, even the odds of a recession.

• Speaker #0

  Makes sense. Those things definitely have an impact on the markets.

• Speaker #1

  Right. But here's the thing. Even after controlling for all those economic variables, the opposing seasonality in treasury and equity returns was still there. It just wouldn't go away.

• Speaker #0

  Okay. So it wasn't just the economy driving these patterns. What else did they look into?

• Speaker #1

  Well, they also considered something called cross-market hedging. This is when investors shift their money between different asset classes to manage risk. So, for example, if the stock market is looking shaky, investors might sell some of their stocks and buy treasuries to play it safe.

• Speaker #0

  Ah, I see. So they were wondering if maybe investors were just moving their money back and forth between stocks and bonds based on their overall risk appetite.

• Speaker #1

  Exactly. But when they tested this theory, they found that while cross-market hedging did play a small role, it couldn't fully explain the magnitude of those seasonal swings in treasury returns.

• Speaker #0

  So we've ruled out the economy in cross-market hedging. What else did they throw at this puzzle?

• Speaker #1

  Well, they even looked at investor sentiment. You know, how some days everyone seems to be feeling super bullish and other days it's all doom and gloom. They wanted to see if those overall mood swings in the market could be causing these opposing seasonalities.

• Speaker #0

  And did they find a connection?

• Speaker #1

  Surprisingly, no. While investor sentiment does influence market movements, it didn't fully explain this specific pattern either. Wow.

• Speaker #0

  Yeah. They really went above and beyond to test all these possibilities. But I'm guessing they didn't stop there, did they?

• Speaker #1

  Nope. They even brought out the big guns. They tested some of the most well-established asset pricing models out there. including the famous Fama-French three-factor model for equities.

• Speaker #0

  Hold on. For our listeners who aren't finance nerds, can you give us a quick Eli-5 on what these models are all about?

• Speaker #1

  Sure thing. Basically, these models try to explain asset returns based on different factors, like market risk, company size, value versus growth stocks, and so on. They're like the gold standard for understanding what drives market movements.

• Speaker #0

  Got it. So they were testing to see if these fundamental factors could explain those weird seasonal patterns.

• Speaker #1

  Exactly. And the results were pretty fascinating. While some of these factors were significant in explaining overall market returns, they still couldn't fully account for those opposing seasonalities.

• Speaker #0

  So even after throwing everything but the kitchen sink at this problem, they still couldn't explain it away?

• Speaker #1

  Nope. And here's the kicker. Even after adding all these other variables into the mix, that little OPERART variable, the one that measures SAD symptoms, remained statistically significant. It just kept popping up as a key driver.

• Speaker #0

  Okay. I'm starting to see why this SAD theory is so compelling. They tested all these other explanations and none of them fit as well as this idea that people's moods are influencing the bond market.

• Speaker #1

  Right. It might sound strange at first, but remember, at the end of the day, markets are driven by human behavior and human behavior can be pretty unpredictable, especially when our biology and psychology are involved.

• Speaker #0

  So we've got this fascinating link between SAD and market movements and a simple trading rule that seems to capitalize on it. But let's be real for a second. Is this something our listeners can actually use in their algo trading?

• Speaker #1

  That's the million-dollar question, isn't it? The research is definitely compelling, but like any market anomaly, there are a few things to keep in mind.

• Speaker #0

  Okay, so what are the potential pitfalls here?

• Speaker #1

  Well, first off, remember that markets are constantly evolving. What worked in the past might not always work in the future.

• Speaker #0

  That's true. Past performance is never a guarantee of future results.

• Speaker #1

  Exactly. Plus, the researchers... only focused on U.S. treasuries. It's not clear if this same seasonal pattern holds true in other bond markets around the world. More research is definitely needed.

• Speaker #0

  Good point. We don't want our listeners making any rash decisions based on a single study. Anything else we should be Ausha about?

• Speaker #1

  Well, remember we talked about how this trading strategy relies on the fact that not everyone is aware of this SAD linked seasonality. If too many traders start piling into treasuries every October, it could start to distort the market and reduce the profitability of the strategy.

• Speaker #0

  That makes sense. It's like a self-fulfilling prophecy. The more people try to exploit the anomaly, the weaker it becomes.

• Speaker #1

  Exactly. So it's important to approach this with a healthy dose of skepticism and to do your own thorough backtesting before incorporating it into your algo trading.

• Speaker #0

  Speaking of backtesting, are there any other practical considerations our listeners should keep in mind?

• Speaker #1

  Definitely. You always have to factor in things like transaction costs and slippage. And if Of course. Yeah. There's always the risk of unforeseen events that could throw a wrench in your plans.

• Speaker #0

  True. Black swan events can happen anytime. And those can really mess with even the most well-thought-out strategies. But even with those caveats, this SAD-linked seasonality seems like a pretty promising avenue for algo trading, right?

• Speaker #1

  Absolutely. It opens up a whole new way of thinking about market behavior. Instead of just focusing on economic indicators and technical signals, we can start incorporating behavioral facts. factors like mood and sentiment into our models.

• Speaker #0

  That's what I find so exciting about this research. It challenges the traditional view of markets as perfectly rational and efficient.

• Speaker #1

  Right. It reminds us that markets are made up of people. And people are complex, emotional beings. Our decisions aren't always driven by pure logic.

• Speaker #0

  Exactly. And sometimes those seemingly irrational decisions can create predictable patterns that we can exploit in our trading.

• Speaker #1

  So to answer your question from earlier, yes, I think this SAD-linked seasonality is something our listeners can definitely use in their algo trading. But they need to approach it carefully, do their due diligence, and always remember that no strategy is foolproof.

• Speaker #0

  Great advice. It sounds like there's a real opportunity here for those willing to dive deep and explore this further.

• Speaker #1

  Absolutely. And who knows what other fascinating behavioral anomalies are lurking out there in the market, just waiting to be discovered.

• Speaker #0

  That's what keeps things interesting. 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.