An algorithmic testing facility offered by exchange operator Nasdaq OMX aims to let buy-side firms measure the impact of low-latency trading on their own algorithmic trading strategies while also reducing glitches.
The Algo Test Facility will be jointly launched by Nasdaq OMX and Thesys Technologies, the infrastructure affiliate of financial technology firm Tradeworx. It will launch in Q1 2014, initially letting users test equity algos on Nasdaq market data, before extending to all major US equity exchanges.
Although chiefly a tool to mitigate risks associated with algo trading, the system will market participants test execution performance.
Manjor Narang, CEO of Tradeworx, said buy-side firms would be keen to use the system to measure the impact of high-frequency trading (HFT) on their algo strategies.
In meetings with buy-side firms, Narang said trading heads were concerned over low-latency activity in the market. Specifically they wanted to know what effect this activity had on institutional investors’ algo execution.
“This platform has the ability to quantitatively answer that question,” he said “It’s designed to show you how much latency affects the performance of your trades.”
The system is also part of efforts by Nasdaq to address the growing prevalence of market disruptions associated with algorithmic trading in light of high-profile market events such as last year’s Knight Securities market-making glitch, which ultimately wiped US$440 million from the firm’s share price.
Nasdaq itself was forced to halt trading for three hours in August due to a systems glitch and last year its Facebook IPO was subject to execution issues that forced orders to be cancelled.
Stacie Swantstrom, head of access services at Nasdaq, said the algo testing facility would help address the technological complexity of algo trading to reduce the likelihood of such errors. She added that industry-led solutions would be benefit the market more than more regulation.
“This is the first step by Nasdaq OMX to improve testing for the larger US equities market ecosystem and proactively get ahead of potential regulatory mandates to deal with issues linked to algorithmic trading,” Swanstrom said, adding the market operator would announce other initiatives in the near future.
The ability to test algos before they are used within the market will also help asset managers and other participants deal with the growing complexity and fragmentation of the US equity market.
Tradeworx’s Narang, whose firm has operated the testing facility technology since 2009, said Regulation National Market System (Reg NMS) had caused a significant increase in market complexity, leading to more trading errors.
“Some of the major glitches that occur, such as quote stuffing (submitting then withdrawing quotes that competitors have to process), actually arise out of inadvertent behaviour that is introduced by compliance with Reg NMS,” he said.
“We’ve tried to introduce as many elements as possible from Reg NMS to simulate the entire market structure, not just the matching engines of exchanges to include the market’s overarching regulatory framework.”