BUZZ US Sentiment Leaders ETF
Artificial intelligence and machine learning is the last, best hope for active management. The idea is that infinitely powerful computers, ruthlessly trained on the market, will be able to deliver excess returns in ways no humans could.
But it’s not just an idea; it’s already happening.
The buzz us sentiment leaders ETF (BUZ) uses natural language processing and artificial intelligence techniques to scan the market for real-time sentiment around leading U.S. large-cap securities. developed by hedge fund experts, it’s a true AI ETF that’s handily outperformed the s&p 500 since it launched in April 2016. the most exciting news? according to Jamie Wise, founder and CEO of buzz indexes, it’s only going to get better.
Matt Hougan, CEO, Inside ETFs: Why did you develop BUZ?
Jamie Wise, Founder and CEO, Buzz Indexes: We’ve always known that sentiment impacts stock price returns. For decades, we’ve relied on proxies to try to measure sentiment: investor surveys, market-based indicators, investor flows, etc. Each had its flaws. The failures led many to believe that sentiment couldn’t deliver; indeed, some said it was a contrarian indicator to stock price performance.
Our view was that, with the explosion in activity of people talking about stocks online, we had a new opportunity. We now, for the first time, had people independently volunteering their views on the prospects of different securities. If we could apply the latest technologies to listen to those conversations, we could understand sentiment down to the individual stock level on a real-time basis, and use that information as a predictive indicator of returns.
How does that practically work? Walk me through the process from me tweeting about Tesla to you building a portfolio.
The general view is that no one expert—whether that’s a research analyst, media personality or star portfolio manager—has the absolute authority to know what’s going to happen with a stock.
The wisdom of crowds suggests that if you have a diverse group of people who are all independently talking about a similar topic, and if they have an incentive to tell the truth (and people do online), then the consensus view of that crowd will typically be more accurate than any one “expert” within that crowd. With BUZ, we’re measuring the sentiment expressed in the conversations people are having about stocks and their investment portfolios. Our view is that those sentiments are more predictive of stock price returns than the opinions of any individual expert.
Which stocks do you look at?
We focus on large-cap U.S. equities, for a few reasons. First, doing so results in a portfolio of stocks that’s familiar to investors. Second, it creates a portfolio that’s both investable and liquid. Third, by focusing on large-caps, you reduce the chances of bad actors—people who are deliberately engaging in online stock promotion. Large-cap equities have so much conversation around them that promotional activity is more easily recognized, filtered out and unlikely to influence overall sentiment.
Beyond that—and this is unique to BUZ—we focus only on those stocks that have exhibited the highest degree of consistency and diversity in their conversation online over the past year. It’s much more significant to us if we note a change in sentiment for a stock that has 50,000 people consistently mentioning it than if we saw a change in sentiment for a stock that only has 50 people talking about it.
Right now, roughly 250 names meet our criteria for the most-discussed large-cap U.S. equities. We set this as our eligible investment universe. When we first launched the index in 2016, that investable universe was just 150 names. But because more and more people are talking about stocks online each day, that universe has expanded and is only continuing to grow.
How do you build the portfolio?
We look at our eligible universe of stocks and rank them monthly on sentiment from highest to lowest. We include the 75 most positively talked-about names in the index, with a 3% maximum weight to make sure we’re sufficiently diversified.
Where does BUZ fit in an investor’s portfolio?
We think of it as a large-cap replacement and a complement to any traditional beta you have. It’s a rules-based approach to active management.
How has it performed?
We’ve beaten the market as a whole since inception, and that gap has increased recently. We think there are two reasons for our strong recent performance.
First, as the eligible investment universe expands, the opportunities to source additional stocks for alpha generation increase.
Second, the model gets smarter over time. That’s one of the hallmarks of AI and machine learning. With more data and experience, the model is able to continually refine and adapt to changing trends, delivering more relevant insights with better accuracy.
When does BUZ do particularly well, and when will it lag the market?
We’ve found that the index does a nice job of outperforming the market when markets are positive, and holds its own in falling markets. That’s the asymmetric return we’re looking for, and it makes sense: If you focus on companies that people are most positive about, they should be cushioned in falling markets, as people should want to buy more as they get cheaper.
If you had to describe BUZ in one paragraph, what would it be?
BUZ lets you access the same leading- edge investment insights that are currently being deployed by the world’s leading quantitative hedge funds. Advancements in artificial intelligence and machine learning, trained to the huge amount of online stock-specific discussion, means you can finally harness the sentiment premia that’s always been present in stock prices.