Lucy Nicholson/Reuters

Dmitry Balyasny speaks at the 2018 Milken Conference in Beverly Hills, California.

  • New alternative data companies are popping up everyday, and hedge funds are getting overwhelmed by choice. Some have developed a way to evaluate data more quickly.
  • Two hedge funds, Balyasny and WorldQuant, have built online portals where sellers can plug in their datasets. The fund managers then quickly relay what data they find valuable.
  • While the process saves hedge funds time, some sellers believe that the lack of human touch will cause funds to miss out on valuable datasets.
  • Click here for more BI Prime stories.

Hedge funds are trading on unconventional data like satellite images and corporate air travel trackers to gain a competitive edge, but picking out useful information is getting harder as data vendors multiply. Funds like WorldQuant and Balyasny want to streamline that process.

So they set up online portals - called Data Exchange at WorldQuant and Antenna at Balyasny - to let sellers upload data and get quick feedback on what's valuable for the funds.

"This is a real transparent way to show our vendors 'here's what we are looking for,'" said Carson Boneck, chief data officer for Balyasny.

"You're going to be put at the top of the queue if Antenna identifies something of value."

See more: Hedge funds' secret sauce is obscure data like satellite images. Here's how the people in charge of spending millions on this data find the stuff worth buying.

Hedge funds are fighting tooth and nail for every basis point of performance. And the entire industry - including quant firms and traditional stock-pickers - is planning to spend billions on alternative data, which has sparked an explosion of new providers.

number alternative data providers chart

Shayanne Gal/Business Insider

After a year where the average hedge fund lost money, 2019 has been kinder to hedge fund managers, with the average fund posting a 7.6% return through the end of June, according to Hedge Fund Research. This figure however trails the overall market and has not stopped investors from redeeming tens of billions of dollars this year.

Hedge fund managers have traditionally found interesting data sets through data-buyers who attend conferences like BattleFin, but the volume of new data providers and datasets has forced managers to come up with other approaches. The number of alternative data sellers has nearly quadrupled over the past ten years, up to 412 firms in 2018.

"In sorting through all of these new vendors, we are trying to find a way to increase the probability that we find something our investors would value, and that would be useful for our portfolio managers and quants," Boneck said.

The quicker data can be vetted and integrated, the more useful that data is, portfolio managers say. An investment signal may disappear, or it can be picked up by others who crowd the trade.

"You get a short-term advantage, but then it closes because everyone else catches on," said Dev Kantesaria, founder of $450 million hedge fund Valley Forge Asset Management.

'Data still needs human interaction'

Balyasny's Antenna is a quantitative review of the dataset that "isn't the end-all, be-all," Boneck said, but the program, which started at the beginning of 2018, "covers a lot of the initial bases."

"It certainly makes me more comfortable purchasing data when I know it's passed all of these objective tests."

See more: Hedge-fund managers are overwhelmed by data, and they're turning to an unlikely source: random people on the internet

For some vendors in the still-new alternative data space, this type of review is helpful in pinpointing what their product is worth - especially since many companies are selling something that has never been sold before.

"In terms of pricing in the market, it's really all over the place," said Sarah McKenna, COO of Sequentum, which builds software for managers to web-scrape data. "It's almost easier to have them bid then for us to state a price."

But some vendors are concerned a quant system might not be able to pick up on the intricacies of their datasets like a human would.

"Data still needs human interaction," said Zak Selbert, CEO and founder of Indexica, a company that uses machine-learning to review and analyze transcripts, media reports, public filings, and more.

Selbert said he understands why hedge funds want to streamline the data-review process, but warns that "a lot of funds would miss out on a lot of good data, if this spreads."

"My hope is that it's just a screening tool, and not a decider. Because if it's the decider, people will only ingest data that has this statistically predictive, immediate look to it, and not dive deeper into others."

See more: Pricey data, slashed fees, and poor returns are hurting hedge funds' margins -and some are getting in the business of helping their rivals

Boneck understands Selbert's and others' concerns, and said that "we're always going to need to go deeper into data."

Balyasny at the moment has no plans to white-label the Antenna algorithm for general use by other funds. There has been a growth in third parties that are running platforms with pre-vetted data vendors on it like Bloomberg and BattleFin, which recently rolled out a service called Ensemble.

This step will help clean up the space, Boneck said, but won't replace the vetting done by Balyasny and others.

"We're always going to need to test it ourselves."

NOW WATCH: An animated map shows every total solar eclipse around the world until 2040

Add Comment
Keep scrolling for next story

Deutsche Bank just nabbed AQR's head of technology to help lead a $15 billion push into digital with a focus on the cloud

Loading next story
Copyright © 2015 Bennett, Coleman & Co. Ltd.
All rights reserved.