The Raleigh, North Carolina-based technology company, which carries out market microstructure model development for institutional clients, has spent three years developing the project, which is referred to simply as ‘the platform’.
It will be targeted at smaller institutional clients, also known as Exempt Contract Participants (ECP), such as commodity trading advisers, which are either firms or individual investors with more than $10 million under management. According to UGS, investors of such a size have not had access to such technology in the past because they were perceived as being too small to trade algos via the large FX banks.
UGS has also added features to its black-box trading model that have not been part of most algo products in the past, claims Elliott Shifman, the company’s president, who developed the platform. For instance, it offers advanced analytics, which include real-time calculation of the order-book skew and commitment of traders’ reports, which have typically only been produced in an end-of-day format.
“Many of the algorithms we’ve seen in the marketplace have been based on functions of price. So whether it’s a momentum model or a mean reversion model, they’re functions of price, not functions of the components behind price,” Shifman tells EuromoneyFXNews. “To our knowledge, [our platform] is unique. Certainly, with the details we’re showing in the analytics, we’re not aware that anyone else is providing that real-time data and openness.”
It is claimed that the UGS platform can be installed in just a few hours instead of days and allows the currency manager full control of strategy, because features allow the manager to see what the algo program is doing.
“In black-box technology, by definition, a customer signs up for a particular algo, and all they see are trades after they’ve been executed,” says Shifman. “Our algo provides the analytics and all sub-analytics of what is behind the algorithm. Being in control means that parameters can be customized for an individual ECP.”
The UGS platform is based on a medium-frequency model rather than a high-frequency one, meaning that trade duration is measured in minutes, not milliseconds, unlike many of the other models that are currently available.
Millisecond algo trading makes thousands of trades, each with tiny profits, whereas UGS’s algo attempts to profit from 50-60 more profitable trades. For example, programs that trade on milliseconds might make .2 of a pip, while a program with medium-frequency duration will look to profit by several pips per trade.