Buy low, sell high Tesla, this artificial intelligence ETF is more accurate than “Sister Wood”

Using AI algorithms to select stocks most likely to outperform the broad market

Buy low, sell high Tesla, this artificial intelligence ETF is more accurate than "Sister Wood"

Through skilled grasp of Tesla’s recent trend and other factors, this ETF has created a 53% return in the past 12 months, higher than the 43% return of the S&P 500 over the same period.

This year’s Tesla stock price trend seems like a roller coaster, even Ark Investment CEO Cathy Wood (Cathie Wood) such a well-known investors are inevitably hit. However, a small ETF that uses artificial intelligence (AI) stock-picking algorithms has achieved several successful buy-and-sell transactions over the past year by buying during the downturn and selling shortly before the drop.

The $27 million ETF, called the QRAFT AI-Enhanced U.S. Large Cap Momentum ETF (AMOM), was launched in 2019 by Qraft Technologies, a South Korean startup asset management firm. The fund does this by first screening 300 large-cap U.S. stocks with the strongest price action and then using AI algorithms to select the 50 stocks most likely to outperform the broader market.

From early 2020 to the end of August of that year, Tesla’s stock price soared nearly 500%, and the ETF’s AI recommended selling all of Tesla’s shares, which fell 22% in the two months after that.

In late November 2020, this ETF’s AI recommended re-buying Tesla shares and added to the position in December 2020 and January 2021, respectively. In those months, Tesla’s stock price jumped 98%, and at the end of January 2021, the ETF sold all of its holdings in Tesla based on the AI’s recommendations. Since then, Tesla shares have fallen more than 30% from their all-time highs.

By skillfully capturing Tesla’s recent moves, among other factors, this ETF has generated a 53% return over the past 12 months, outperforming the 43% return of the S&P 500 over the same period. In the ETF’s latest rebalancing at the end of April, AI once again included Tesla in its portfolio, indicating a belief that Tesla’s stock price would rally. Whether this prediction will be fulfilled again remains to be seen. After the ETF included Tesla again, Tesla shares fell 21%, but rose 4% in trading on May 20.

AI has become a useful tool in different ways for many institutional investors and quantitative investment institutions. Computers can track stock market price trends, access public company releases and earnings reports, digest and interpret economic data in the first place, and even analyze public sentiment through millions of social media posts.

This information is fed into the algorithm, which takes all factors into account and in turn picks stocks that have a high probability of outperforming the broader market. In contrast to rule-based models, the most important feature of AI models is the ability to learn from their own successes and mistakes through feedback loops, just like humans do.

Although AI models can process large amounts of data in a few seconds, they are not a panacea, according to Geeseok Oh, head of Qraft Technologies’ Asia Pacific office, who says that AI is not able to observe social trends or understand technological innovations the way humans do, making it unsuitable for long-term bets. framework to find the stocks that are most likely to outperform the broader market.

While hedge funds and professional traders have been using AI to improve performance, there are very few retail-oriented products run purely by AI models. in 2017, ETFMG launched its $145 million AI Powered Equity ETF (AIEQ), the first active ETF to fully leverage AI stock selection, using IBM’s Watson machine learning system. According to ETFMG, the system is “the equivalent of a team of 1,000 research analysts, traders and quantitative analysts working around the clock.

Qraft Technologies is a newcomer to the space, but the strategy of using AI to pick stocks has been the company’s focus from the start. The company’s founder, Marcus Kim, has a background in software engineering and algorithmic trading, and Geeseok Oh, head of Qraft Technologies’ Asia-Pacific office, told Barron’s that the company’s AI models are not built from scratch in-house using a general-purpose system like IBM’s Watson. In addition to this ETF, the company offers other funds that specialize in high-dividend and value stocks, all of which are built and managed by AI.

However, AI-driven ETFs are not yet available on a wide scale, and investors have good reason to stay on the sidelines. Unlike thematic ETFs, it is difficult to explain how AI algorithms work, which often involve abstract modeling and a lot of terminology, and the data processing is a bit of a “black box” that does not always have clear rules. Very often, even the data scientists who build the models cannot explain why the computer made a certain decision.

There is also the problem of “input garbage, output garbage,” which means that if the input data is flawed, the output will be meaningless. Many computer scientists, including those at Qraft Technologies, believe that social media sentiment is more of a noise than a signal, and Geeseok Oh says, “It may be good for marketing, but I’m not so sure how much it means for fund performance itself.”

The lack of past performance records is another obstacle. Like any human fund manager, AI algorithms need some time to prove they have the ability to perform strongly enough to gain investor trust, and the first few years of that process can be very difficult.

ETFMG’s AI funds underperformed the S&P 500 in 2018, matched the index in 2019 and beat it in 2020, and Qraft Technologies’ longest-running ETF just passed its second anniversary, and its returns have performed well so far. As a result, Geeseok Oh said he sees investor interest building. Still, these funds are small and still have a long way to go in the future.

Posted by:CoinYuppie,Reprinted with attribution to:
Coinyuppie is an open information publishing platform, all information provided is not related to the views and positions of coinyuppie, and does not constitute any investment and financial advice. Users are expected to carefully screen and prevent risks.

Like (0)
Donate Buy me a coffee Buy me a coffee
Previous 2021-05-30 22:25
Next 2021-05-30 22:30

Related articles