Eight months ago, Bloomberg reported that machine learning, according to Gartner Inc.’s hype cycle for new technologies, was at the “peak of inflated expectations” and heading to the “trough of disillusionment.”
Around the same time a number of A.I.-driven investment funds and ETFs began to pop up, seemingly confirming the hype cycle.
In October 2017, the first Artificial Intelligence-driven ETF was launched:
“EquBot AI Technology with Watson has the ability to mimic an army of equity research analysts working around the clock, 365 days a year, while removing human error and bias from the process.
And while hedge funds like Bridgewater have successfully automated many of their trading strategies and Renaissance Technologies, a quant fund founded by former military code-breaker Jim Simons, has been a pioneer in using machine-learning techniques for decades while building an enviable investing track record; probably the best known of these new funds was a push by Paul Tudor Jones to build an algorithm that would mimic his analytical process
Back in the 1990s, Paul Tudor Jones assigned a team of coders to a project dubbed “Paul in a Box.” The effort sought to break down the DNA of the hedge fund manager’s trading – how he sizes up markets and generates ideas – to train a computer to do the same.
The code created then was upgraded many times and is still used at his firm, Tudor Investment Corp. But it never took over.
Again and again, programmers had to feed in new types of data to mimic the changing price signals that Jones, famous for predicting the Black Monday market crash 30 years ago, zeroed in on, according to people with knowledge of the project. Even then, the machine couldn’t capture intangibles like his gut instincts and conviction, as well as the market’s uncertainties.
Ultimately, Jones remains the final decision-maker for trades – not the box.
Since then, things have not quite worked out how Gartner – and likely numerous human traders – would have hoped.
First things first, the AI ETF – that was designed to end the careers of 100s of analysts – has outperformed the market dramatically since inception…
Whether this is pure momo-chasing in the algo is unclear – but is notable that on the downside swing in February, the ETF did not underperform dramatically.
But it is the success of Paul Tudor Jones’ AI Fund that stands out, as Bloomberg reports, Tudor Investment Corp.’s Dharmesh Maniyar posted a 7 percent gain last month in his new fund that uses machine-learning algorithms to help make macro trades.
The performance brings returns for his Tudor Maniyar Macro fund this year to about 13 percent, according to people with knowledge of the matter – well ahead of the S&P 500. The fund reportedly made money on European fixed income and global currencies in May.
Tudor had raised about $300 million for Maniyar’s fund, which started trading in October and is the firm’s only macro fund run by a sole manager.
But interestingly, Maniyar, who has a doctorate in machine learning from the U.K.’s Aston University, got off to a poor start, slumping 8 percent in the first three months of trading last year, one of the people said – which mimics the same major underperformance seen early on in the AI ETF above.
Just like humans, perhaps the machines just needed to get used to the new desk and new normal.
For now, it appears that just as retail investors piling into passive investment vehicles is killing the active manager, so the same active manager is now being trumped by Johnny 5 – time to start learning a new skill (like horse and buggy repair?).
via RSS https://ift.tt/2M5lIjm Tyler Durden