Transcription

• Speaker #0

  Hello, welcome back to Papers with Backtest podcast. Today we dive into another auto trading research paper exploring a strategy connected to short interest, specifically the surprising upside of low short interest stocks. You shared this paper because you're interested in finding new signals for undervaluation. What this research shows is that low short interest, especially in heavily traded stocks, could be such a signal.

• Speaker #1

  What's fascinating here is that it really challenges how we typically view short interest. High short interest is often seen as a warning sign, but this research suggests that we might be missing valuable information by ignoring the stocks that short sellers are avoiding.

• Speaker #0

  The researchers found these low short interest stocks consistently outperformed the market. And this wasn't just a short-term blip. The effect lasted for at least six months, making it potentially valuable for longer-term strategies too.

• Speaker #1

  Exactly. And the paper we're looking at is from 2009, titled The Good News in Short Interest by Bomer, Hussar, and Jordan. They looked at data from 1988 to 2005 across the NYSE, Amex, and Nasdaq. So pretty extensive data set.

• Speaker #0

  Okay, so let's dive into the specifics. They used the short interest ratio, or SIR, to measure short interest. But we know SIR alone can be misleading. What are your thoughts on refining this signal? Do you prefer days to cover or maybe factor in short interest relative to the float?

• Speaker #1

  I think those are all valid considerations. The key is to find a metric that truly reflects the scarcity of shares available for shorting. Days to cover. Which factors in trading volume could be a good starting point. And considering short interest relative to the float helps us understand how much of the available stock is being shorted.

• Speaker #0

  The researchers grouped stocks into different SAR percentiles and built portfolios based on those rankings. And this is where things get really interesting. The portfolio with the lowest SR, meaning the least short interest. generated an average monthly return of 2.1%. That's significantly higher than the returns observed for the highest SR percentile portfolios, which were actually slightly negative.

• Speaker #1

  You're right. Those returns are pretty impressive. And remember, they persisted for at least six months. This isn't just a quick arbitrage opportunity. There seems to be a more persistent mispricing happening.

• Speaker #0

  I'm curious about why this might be happening. The researchers offered two possible explanations. The first is an information gap. Short sellers, who are often considered sophisticated investors, might be avoiding certain heavily traded stocks because they simply don't see an opportunity to profit.

• Speaker #1

  Right. They might be seeing potential that the broader market hasn't fully recognized yet. So by tapping into this low short interest signal, you might be getting a sneak peek into those potentially undervalued gems.

• Speaker #0

  The second explanation is a bit more mechanical, and it has to do with the constraints of short selling itself. particularly for smaller companies, it can be challenging to borrow shares to short.

• Speaker #1

  Exactly. Even if a short seller believes a stock is overvalued, if it's difficult or expensive to borrow shares, they might simply stay away, leading to artificially low short interest.

• Speaker #0

  So it's not necessarily a reflection of true investor sentiment, but rather a limitation in the short selling mechanism. Given that, I'm curious how much of this outperformance is driven by genuinely undervalued stocks versus those that are just hard to short?

• Speaker #1

  It's a fascinating question, and one that the paper doesn't directly answer. But I think it's safe to assume that both factors are at play. The key is to find ways to disentangle those two effects. Perhaps by focusing on stocks with a high degree of institutional ownership, which are generally easier to borrow.

• Speaker #0

  So we might want to layer in some additional filters to refine our selection process. For example, looking at short interest as a percentage of the free float, or focusing on stocks with high share turnover. might help us identify those genuinely undervalued opportunities.

• Speaker #1

  Those are all excellent points, and it highlights the importance of not just blindly following a single metric, but rather understanding the underlying dynamics and refining our approach accordingly.

• Speaker #0

  Okay, so we've talked about the potential for uncovering undervalued stocks, but let's get into the meat of how this might translate into a trading strategy. The paper backtested a very simple long-only strategy based on this short interest effect. They bought the lowest SR percentile each month. Rebalancing monthly and holding the portfolio equally weighted.

• Speaker #1

  Right. And that's where the results become truly eye-catching. The backtest showed that this simple, low short interest portfolio generated an impressive annualized return of 26.8%.

• Speaker #0

  Hold on. Almost 27% per year. You mentioned diversification. But given the short selling constraints, wouldn't focusing on a niche of low short interest stocks inherently limit diversification opportunities?

• Speaker #1

  That's a valid concern. And it's something to be mindful of, however. Keep in mind that the researchers tested this across a large universe of stocks from different exchanges. So while there might be some limitations, there's still potential to build a reasonably diversified portfolio within this low short interest segment.

• Speaker #0

  That makes sense. But to play devil's advocate here. These impressive returns are based on historical data. The market has evolved significantly since 2009. How confident are you that this strategy would still work today?

• Speaker #1

  That's a really important question and one we need to approach with a healthy dose of skepticism. Of course, past performance is never a guarantee of future returns. But the fact that this effect was observed over such a long period and across different market environments does suggest that there might be a persistent behavioral element.

• Speaker #0

  Right. It's not just a fluke or a product of a specific market regime. It's possible that short sellers, with their in-depth research, are consistently identifying stocks with hidden potential. And their avoidance of these stocks creates an opportunity for other investors.

• Speaker #1

  Exactly. And it highlights the potential for uncovering alpha by looking where others aren't. The key is to adapt this strategy to the current market dynamics. For example, incorporating factors like short interest to the float. days to cover, or even sentiment analysis could help refine the signal and potentially enhance its effectiveness. I think incorporating those additional filters is crucial. It's not just about finding stocks with low short interest. It's about understanding why the short interest is low and identifying those situations where it truly signals undervaluation.

• Speaker #0

  So we're essentially trying to get into the minds of the short sellers, figure out what they're seeing, and potentially front run their eventual realization of that value.

• Speaker #1

  That's a good way to put it. We're looking for those situations where the short sellers are right about the upside potential, but wrong about the timing.

• Speaker #0

  OK, so we've got our filters in place. We've identified a potential pool of undervalued stocks. What's the next step? How do we actually turn this insight into a trading strategy?

• Speaker #1

  Well, the paper provides a good starting point with their simple long only strategy. But I think there's room for improvement, for example, instead of simply buying the lowest SR percentile. We could consider a tiered approach.

• Speaker #0

  What do you mean by a tiered approach?

• Speaker #1

  We could divide the low short interest universe into further subgroups based on additional criteria, like days to cover or short interest relative to the float. And we could allocate different weights to each subgroup based on our conviction level.

• Speaker #0

  So we're essentially adding layers of sophistication to the strategy, refining our selection process, and potentially enhancing our risk-adjusted returns.

• Speaker #1

  Exactly. And we could also consider incorporating other factors into our trading rules, such as momentum indicators or fundamental analysis, to further improve our odds of success.

• Speaker #0

  It sounds like building a robust trading strategy around this low, short interest signal requires a lot of thought and experimentation.

• Speaker #1

  It does. There's no one-size-fits-all solution. The optimal approach will depend on your specific investment goals, risk tolerance, and time horizon.

• Speaker #0

  One thing that struck me about the paper's backtest results was that they used an equal weighted portfolio. Wouldn't a market cap weighted approach be more realistic, given that most investors wouldn't be allocating equal capital to every stock?

• Speaker #1

  That's an interesting point. The authors actually tested both equal weighted and value weighted portfolios. And interestingly, the equal weighted portfolio consistently outperformed. Wait,

• Speaker #0

  that's counterintuitive. Why do you think that is?

• Speaker #1

  It could be because the undervaluation effect is more pronounced in smaller stocks, which would receive a greater weight in an equal weighted portfolio. But it's also possible that this is just a quirk of the historical data. I think it would be worthwhile to explore different weighting schemes to see which one performs best under different market conditions.

• Speaker #0

  That's a good idea. So we've got this potential strategy based on low short interest. But before we go live with it, we need to backtest it thoroughly to make sure it's robust and can handle different market scenarios. What are some key considerations when backtesting a strategy like this?

• Speaker #1

  One of the most important things is to use a realistic data set. We need data that accurately reflects the market conditions we're likely to encounter in live trading. This includes data on stock prices, trading volumes, short interest, and any other factors we're incorporating into our strategy.

• Speaker #0

  And we need to be mindful of backtesting bias, right? We want to make sure we're not just curve-fitting our strategy to historical data.

• Speaker #1

  Exactly. There are several techniques we can use to mitigate backtesting bias, such as out-of-sample testing and walk-forward analysis. Out-of-sample testing involves reserving a portion of our data for testing purposes, so we can evaluate how well our strategy performs on data it's never seen before. Walk-forward analysis simulates real-time trading by gradually expanding our testing window as we move forward in time.

• Speaker #0

  So we're essentially trying to create a more realistic back-testing environment that reflects the dynamic nature of the market.

• Speaker #1

  Precisely. And in addition to those techniques, we also need to be careful about the parameters we're optimizing. We want to avoid over-optimizing our strategy to fit the historical data too perfectly, because that can lead to poor performance in live trading.

• Speaker #0

  Right. It's all about finding that sweet spot between optimization and robustness.

• Speaker #1

  Exactly. And another important consideration is the choice of performance metrics. We need metrics that accurately reflect the risk and return characteristics of our strategy.

• Speaker #0

  like the Sharpe ratio, which measures risk-adjusted return or maximum drawdown, which tells us the largest potential loss we could experience.

• Speaker #1

  Those are both good examples. We might also want to consider metrics like win-loss ratio, average trade duration, and volatility.

• Speaker #0

  The goal is to get a holistic view of our strategy's performance, not just focus on one single metric.

• Speaker #1

  Absolutely. And we also need to consider the transaction costs associated with trading, such as commissions and slippage. These costs can eat into our profits. So we need to factor them into our backtesting and make sure our strategy is still profitable after accounting for these costs.

• Speaker #0

  That's a good point. Transaction costs are often overlooked in backtesting, but they can make a big difference in the real world.

• Speaker #1

  They can. So we've backtested our strategy, we're confident in its robustness, and we're ready to go live. What are some key considerations for implementing this strategy in real-time trading?

• Speaker #0

  Well, first and foremost, we need a reliable source of short interest data. We need data that's accurate, timely. and preferably covers a wide universe of stocks.

• Speaker #1

  And we need to make sure our trading platform can handle the specific requirements of our strategy. For example, if we're incorporating real-time short interest data into our trading signals, we need a platform that can process that data quickly and efficiently.

• Speaker #0

  And we need a robust risk management system in place to protect our capital. This includes setting appropriate position sizes, stop-loss orders, and diversification limits.

• Speaker #1

  Those are all essential components of any sound trading strategy. And it's important to remember that even the best back-tested strategies can experience drawdowns and losses.

• Speaker #0

  So we need to have a plan for how we'll handle those inevitable setbacks.

• Speaker #1

  Absolutely. We need to be mentally prepared for the ups and downs of the market. And we need to stick to our plan even when things aren't going our way.

• Speaker #0

  That's easier said than done, but it's crucial for long-term success in trading.

• Speaker #1

  It is. And it's also important to continuously monitor our strategy's performance and make adjustments as needed. The market is constantly evolving. So we need to be adaptive and willing to refine our approach over time.

• Speaker #0

  This low short interest strategy seems incredibly promising, but it also highlights the importance of combining thorough research, rigorous backtesting and a disciplined trading approach.

• Speaker #1

  I couldn't agree more. It's not just about finding a magic formula. It's about understanding the market dynamics, managing risk effectively and adapting to changing conditions.

• Speaker #0

  That's really what separates successful traders from the rest.

• Speaker #1

  It is. And remember. This strategy is just one piece of the puzzle. It's not a silver bullet. The key to success in trading is to develop a diverse toolkit of strategies, each with its own strengths and weaknesses, and to use those strategies strategically based on the prevailing market conditions.

• Speaker #0

  It's all about having a holistic approach to trading and continuously learning and evolving. It really is a fascinating journey, this whole process of exploring a new research idea, backtesting it, refining it, and eventually putting it into practice. It's like we're building a trading machine piece by piece.

• Speaker #1

  That's a great analogy. And just like any complex machine, each component needs to be carefully calibrated and tested to ensure optimal performance.

• Speaker #0

  I want to touch on something that often gets overlooked when discussing quantitative strategies, the psychological aspect of trading. Even with a robust, back-tested system, emotions can creep in and sabotage our best intentions. How do you think traders can stay mentally disciplined? Especially when dealing with a strategy like this, that might involve periods of underperformance.

• Speaker #1

  That's an incredibly important point. Trading psychology is often the missing link between a well-designed strategy and consistent profitability. One thing that helps me is to view trading as a process, not a series of individual bets.

• Speaker #0

  It's about trusting the system you've built, rather than getting caught up in the emotional swings of each trade.

• Speaker #1

  Exactly. And it's also crucial to have a clear risk management plan and stick to it religiously. Knowing you have safeguards in place can help reduce anxiety during inevitable market fluctuations.

• Speaker #0

  Right. It takes the emotion out of the decision-making process. So if our listeners are intrigued by this low short interest strategy and want to explore further, what resources or tools would you recommend?

• Speaker #1

  There are a lot of great resources available. For data, there are traditional providers like Bloomberg, FactSet, and Refinitiv, but also specialized vendors who focus specifically on short interest data. They often provide more granular and timely information, which can be crucial for this kind of strategy.

• Speaker #0

  And for actually building and backtesting this strategy, platforms like Quantopian, Quanticonnect, and TradingView are all popular choices. Or for those who are more code-inclined, developing custom algorithms and backtesting environments in Python or R is always an option.

• Speaker #1

  Exactly. The key is to find the tools that best suit your level of expertise and the specific needs of your strategy. And don't forget about the wealth of free resources out there. Academic papers, blogs, online forums. There's a vibrant community of quant traders constantly sharing ideas and insights.

• Speaker #0

  It's a fantastic time to be exploring these kinds of strategies. The tools and data are more accessible than ever before.

• Speaker #1

  It is. And the potential for uncovering hidden alpha is still out there, for those willing to do the research and put in the work.

• Speaker #0

  So to recap, low short interest, especially in heavily traded stocks, could be a valuable signal for identifying undervalued opportunities. The research suggests this effect is persistent, but it's crucial to refine the signal with additional filters, backtest thoroughly, and manage risk carefully.

• Speaker #1

  I would add that it's also crucial to be aware of the limitations of any single strategy. This low short interest approach is not a silver bullet. It's just one tool in a trader's arsenal. Diversification, continuous learning, and a disciplined approach to trading are all essential for long-term success.

• Speaker #0

  Excellent advice. This has been a fascinating deep dive into the world of low short interest trading. Thank you for sharing your expertise and insights.

• Speaker #1

  It was my pleasure. I always enjoy exploring these unconventional ideas and challenging the status quo.

• 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!