BestEx Research Adds US Equities to Pre-Trade Analytics Suite with Predictive Transaction Cost Model Accounting for Speed and Time of Day

EmitenTrust.com The firm's transaction cost models are designed to support portfolio construction in addition to trading strategy optimization and performance evaluation, and are available through clients' existing AI assistants, REST API, and a point-and-click interface.

STAMFORD, Conn., Sept. 30, 2026 /PRNewswire/ -- BestEx Research Group LLC, an independent provider of execution algorithms, trading technology, and trade analytics for global equities and futures, today announced the launch of the Pulse Market Impact Model for US Equities, extending the firm's pre-trade analytics suite. The model delivers symbol-level transaction cost estimates that can be customized by order size, execution speed, duration, and time of day. It is designed not just for post-trade TCA, but also for trading strategy optimization and portfolio construction. In these applications, inaccurate cost estimates can directly affect position sizing and net returns.

What sets the firm's transaction cost model apart is its ability to capture the interaction among order size, execution speed, and time of day, and its out-of-sample performance across a wide range of trading conditions. Across a wide variety of order-size buckets, liquidity buckets, market conditions, and intraday periods, its predictions were within a fraction of a basis point of realized execution costs.

Traditional transaction cost models typically account for order size and execution speed, but often do not adequately capture time of day. That matters because the same-size order can sometimes be traded faster at one point in the day and still incur lower costs than a more slowly executed order at another point in the day. Size and speed are also highly correlated in historical execution data, making their individual effects difficult to isolate. The model was built using a unique hybrid approach, combining more than one million institutional parent orders with transaction cost sensitivities estimated from TAQ data, and then tested on a full year of executions out of sample. The model's methodology and validation are described transparently in a white paper authored by Hitesh Mittal.

"Portfolio managers and quantitative execution researchers need cost models that are robust enough to rely on for portfolio optimization and execution optimization," said Hitesh Mittal, Founder and CEO of BestEx Research. "That requires more than explaining historical trading costs; the model has to be predictive. If you can accurately estimate the cost of an order before it is traded, and understand how that cost changes with execution speed and time of day, you can use it to select better execution strategies. For portfolio managers, a dependable estimate of trading cost feeds directly into position sizing and portfolio construction. Overestimation of cost can cause attractive positions to be sized too small or rejected altogether, while underestimation can make a trade look more attractive than it really is."

The model is accompanied by four years of historical trading analytics, including minute-by-minute bar data and predictive analytics such as volume, volatility, spread, and depth estimates. The model and its supporting analytics are available in three ways: conversationally through Pulse AI, which connects clients' existing AI assistants to the suite; directly through a REST API; and through AMS One, a point-and-click interface for end-to-end algorithmic trading and analytics. Institutional traders can explore the model and schedule a demo on the firm's website. The model's complete methodology and performance are documented in a research paper available to clients upon request.

Pulse Analytics currently includes cost estimation for US equities and global futures, with global equities cost estimation in development. As the firm announced in August, its AI-ready broker-neutral post-trade transaction cost analysis capability is scheduled for release in early 2027.

About BestEx Research

BestEx Research Group LLC is a provider of sophisticated execution algorithms for equities and futures aimed at reducing trading costs for buy-side managers. The firm's cloud-based Algorithm Management System (AMS One) combines their execution algorithms with a user-friendly dashboard, transaction cost analysis, customization, and automation in the industry's first multi-asset, independent algorithmic execution platform. BestEx Research also offers sell-side firms a seamless, customizable trading solution for their clients with no coding required. For more information on BestEx Research's mission and products, or to request a product demo, visit www.bestexresearch.com.

Media Contact
Mohit Shah
press@bestexresearch.com 

View original content to download multimedia:https://www.prnewswire.com/news-releases/bestex-research-adds-us-equities-to-pre-trade-analytics-suite-with-predictive-transaction-cost-model-accounting-for-speed-and-time-of-day-302893669.html

SOURCE BestEx Research