Stifel Electronic Trading Team has developed an intra-day momentum signal to predict short-term price movement with a high degree of accuracy.
The signal is designed to allow all Stifel algorithms to reduce execution costs and improve spread-capture, independent of the parent strategy, according to John Spensieri, Stifel Head of U.S. Equity Trading.
“Once an algorithm decides it should trade, the signal is given discretion to dictate the optimal order placement strategy in light of real-time stock specific behavior,” he told Traders Magazine.
He said that navigating today’s increasingly fragmented market requires a vigilant approach to deploying execution strategies that preserve alpha and minimize implicit costs.
While many factors influence the success of an execution strategy, order placement and pricing are two of the most critical, he said.
“To optimize order placement and pricing, the ideal strategy would be equipped with knowledge of imminent price changes and their direction, allowing fill-by-fill reductions in execution costs,” he said.
The signal demonstrates consistently high accuracy in back-tests and production, said Spensieri, adding that the duration of the signal is ticker-specific and dependent on quote stability.
“Since the signal responds to market action instead of market expectation during event-driven volatility, it is more robust to external noise than more sentiment-based models,” he said.
“The real-time analysis of trade data is more responsive than the internal GARCH model, which suffers from a characteristic lag,” he added.
According to Spensieri, Stifel’s clientele runs the gamut of asset managers, from mutual funds to quantitative hedge-funds.
“Our clients benefit from the momentum signal by achieving reduced execution costs,” he said.
Spensieri added that Stifel’s U.S. Equity Trading Desk offers a proprietary suite of algorithms, in addition to Portfolio, ETF, Cash, Options, and Convertible Bond Trading.
He said that Stifel’s shift to a proprietary algorithmic product began in 2019 with the implementation of a cloud-based data infrastructure, which provided our foundation for quantitative execution research, and subsequently their algorithm development process.
“We also launched our Stifel X ATS program to facilitate interaction between Stifel retail and institutional equity flows,” he said.
According to Spensieri, Stifel’s algorithmic development process and products focus on optimizing trader experience in providing performance, flexibility, and transparency that reduces unexpected outcomes.
“In turn, the efficiency of our design and development process further strengthens our ability to offer unique solutions to client needs in relatively short time-frames,” he said.
Since Beta deployment of the signal to Stifel’s Liquidity Seeking and Implementation Shortfall algorithms, near-side fills on market orders increased 8 percentage points with a corresponding reduction in far-side fill percentages, Spensieri said.
In addition, spread-adjusted performance on market orders improved 6% with an overall increase in participation rates, he said.
On the back of these positive results, Stifel will deploy the signal to VWAP, TWAP, POV, and SPY Dark Aggregator over the coming months.
Concurrently, Stifel’s Electronic Trading Team will continue to collect feedback from existing algorithms to develop practical solutions that add measurable value to our clients’ execution process.
“Our team is currently testing a new implementation shortfall strategy that utilizes client definable factors to fine-tune the sensitivity of signals within the algorithm,” commented Spensieri.
“This provides clients with more transparency and control with respect to the behavior of the algorithm,” he concluded.
