Nadir Khan was a senior at the Karachi American School in Pakistan when he read Trading to Win by Ari Kiev, the psychiatrist who famously coached traders at Steven Cohen’s hedge fund firm, SAC Capital Advisors. After graduating from Columbia University in 2003, Khan, who collected thousands of pages of stock research data in his spare time, worked as a trader at Cohen’s firm in Connecticut. Three years later he co-founded his own hedge fund firm, which relied on gathering data and drawing correlations between happenings around the world and asset prices.
The hedge fund, Timescape Global Capital Management, closed in 2010, but Khan is now reinventing the data strategy behind it. Using 230 researchers based in Pakistan, he is building a data company called Qineqt that fundamental stock researchers can use to find correlations and patterns that can lead to price moves. Khan is aiming the product at researchers at money management firms, including hedge funds, and Wall Street banks so they can input their investment process and raw information into Qineqt’s data architecture.
As Khan explains, it’s intense and tedious work to get data and information from primary sources. His researchers call stores around the globe to get information on inventory, contact farmers for fertilizer prices, and scout out Chinese ports to find out how many ships are docked. To get access to the necessary number of people to work the phones and systematically collect that kind of granular information, Khan hired experienced people in data processing and equity research in Karachi. Once Qineqt — which will go live with a pilot of five asset managers in December — has its foundation of data, it will use machine learning and other technologies to automate the collection process. Khan points out that many small hedge fund firms and others lose their intellectual capital, along with data developed over years, when star managers walk out the door. Though Qineqt will compete with Bloomberg, Standard & Poor’s, FactSet, and start-ups such as Goldman Sachs Group–backed Kensho Technologies, Khan says his company has the advantage of building a data architecture from scratch in a way that meshes with current thinking and computing power. (Institutional Investor publishes several indexes based on Kensho’s data, and Kensho co-founder and CEO Daniel Nadler writes a column for institutionalinvestor.com.)
Qineqt’s business model centers on forming close partnerships with large money managers that may request, say, 50 data sets in a year and then get exclusive use of the information for a certain period of time after that. The data is then rolled into Qineqt’s central knowledge bank, and other clients can use it. Khan says asset managers are not delivering adequate returns in part because they’re all looking at the same data and crowding into the same trades. “Our platform will provide granular data that has been filtered by a human being,” he says. “They’ll get a clear read of information instead of staring at a screen full of computer-generated data that everyone else is also looking at.”
Take satellite imagery. Khan says in a few years there will be more of this type of data available than most money managers can process on their own. Qineqt will have the ability to tell investors what the satellite images are signaling.
“The value is how it boils down to the earnings power of a company,” he says.
He thinks of Qineqt as a neutral platform for data collection. As a hypothetical example, Khan says the company could negotiate with Uber Technologies for data, something another firm may not be able to do on its own because of cost or other reasons.
Qineqt’s data will be enriched with every money manager that uses the database and adds data and insights.
“For us it’s not what we have right now,” Khan says. “It’s what we’ll have in the future.”
Clearly, Qineqt’s model also depends on security: All managers’ investment strategies are behind firewalls, and no one has access to how any other client has used the data, Khan says.