SOR has been enabling automation in equities asset class for two decades. It’s now poised to enhance the bond market
Increases in fixed income electronic trading volumes have driven enhancements in trading desk operations over the previous five years, making the lives of traders easier in today’s natively digital world. Trading workflow enhancements, however, that automate trading tasks, such as pricing automation and smart order routing (SOR) to different electronic trading venues, have lagged. This has largely been due to the difficulty to aggregate disparate data streams coming from multiple-venues and data-sources in real-time, which is necessary for algorithms to perform automated trading tasks with precision. With the introduction of SOR Artificial Intelligence (AI) algorithms that bridge the gap between data aggregation layer and electronic venues, resulting in a simplified trading process, this is changing rapidly right now.
SOR has been a key component of equity trading workflow for twenty years. By applying AI and algorithms, it examines which venue trades should be routed to for execution and which dealers should be approached, based on past performance and real-time market conditions. SOR will expand the use of algorithms in fixed income trading to create more growth and automation as new execution pathways and scenarios continue to get discovered with AI.
To date, the fixed income industry focus was on automating securities pricing in real-time as a necessary component towards end-to-end trading workflow automation. AI algorithms configured to assist traders with aggregating data sources and automating pricing was the necessary first step. SOR is the natural extension and logical next step as the inter-connecting piece in the trading workflow required to complete end-to-end automation. What’s more, utilizing SOR to automate trade routing and venue selection would free up time and resources for traders to focus on other value-add market monitoring and research activities.
According to Greenwich Associates in their September 2020 report, The Fixed-Income Trading Evolution System: “The time has come to throw out long-held notions of what trading technology can do and how it has been packaged and instead, consider what it should do and what you really need.”
SOR is a step in that direction.
What is SOR?
SOR is an algorithm component of the automated trading system. SOR’s role is to find the best place to execute a trade, based on price and liquidity, across a range of electronic venues and counterparties in the fastest time. By analyzing different offers and placing optimized orders, SOR is a response to the challenge of fragmented data and liquidity in the fixed-income marketplace. SOR algorithms apply machine learning so that each router path is configured and trained on the data record of the previously executed transactions as well as the trading style and bias of the traders it serves.
SOR provides these benefits:
- Automatically accesses a range of venues simultaneously
- Quickly scans and finds the most optimized prices
- Creates a structure and the criteria required to customize algorithms, and;
- Helps track, validate and review data for additional controls and analytics
How it works
The SOR engine determines which venue will be directed to activate the trade, factoring in venue liquidity characteristics and the trader’s specific price momentum and execution cost requirements. The routing logic dynamically adapts based on machine learning of the historical execution records from each specific trading venue and the quality of the available real-time market data at any given moment. The order is then routed through a FIX or custom API gateway. An order route is kept dynamic so that it continues to update existing opportunities and scan for new, better ones as the market conditions change.
SOR algorithms use an ensemble of machine learning models that collate data from multiple sources in real-time and analyze historical and new market data simultaneously. Algorithms determine the size of the trade, execution time, and price, aiming for a volume-weighted or size adjusted best-executable price and trading route.
Why bond trading needs SOR
Currently, there’s a lack of reliable data to guide accurate bond pricing and build credit curves. Traders often have access to only a fraction of data available. The bond market relies heavily on segregated data disseminated between counterparties. This creates fragmented data sets. In addition, most of the existing fixed income data sources do not have enough coverage to provide traders with a view of true liquidity.
Both sell-side dealers and buy-side asset managers are also increasingly relying on AI applications to price bonds in live trading. This includes the consumption of increasing amounts of alternative data and using new methods of fixed income pricing analysis.
Existing technological challenges are in performing data aggregation, mapping, and reconciliation of multiple data streams in real-time so that both historical and current AI modeling optimizations can be performed in a matter of seconds. This is where the application of server-less cloud technologies is proving transformational. AI algorithms that would normally take hours to finish calculations are now processed in seconds, as they are broken down into multiple parallel sequences that serverless cloud infrastructure can compute much faster.
With these recent technological advancements, the natural extension and logical next step at the fixed income trading level is automating trading workflow further with AI-enabled SOR.
AI-Enabled SOR and the Future
Look out for increasing the use of AI algorithms powering various types of SOR protocols that will find new routes and enable new types of trading strategies. This will result in increased liquidity, trading volumes, and ultimately investor returns, raising the bar of market efficiency further. What’s more, end-to-end automation and fully cloud-enabled visualizations will be increasingly used to assist traders in monitoring modeled output and tracking the performance of new trading strategies in an intuitive way.
Going forward, innovation will be driven by API deployment as data can be fed into any internal system or electronic trading venue, making the use of SOR algorithms even more efficient. The marketplace could see the doubling or tripling in the trade execution response speed and overall trading volumes.
1. Quod Financial, “Order Routing and Automated Trading – a new wave of innovation”. Published on July 2019.
2. Renyuan Xu, Isaac Carruthers, “Machine Learning for Limit-Order Routing in Cash Treasury”. Published on June 2018 by Quantitative Brokers.
3. A-Team Insight, “Algorithmic Trading and Smart Order Routing Post-MiFID II”. London TradingTech Summit, February 2019.
4. Michael Kearns, Yuriy Nevmyvaka, “Machine Learning for Market Microstructure and High Frequency Trading”. Published in “High Frequency Trading – New Realities for Traders, Markets and Regulators” on 2013, by Risk Books.
5. Phil Mackintosh, “Routing 201: Some of the Choices an Algo Makes in the Life of an Order”. Published on November 2019 by Nasdaq.
Overbond specializes in custom AI analytics development for clients implementing trade automation workflows, risk management, portfolio modeling, and quantitative finance applications. Overbond supports financial institutions in the AI model development, implementation, and validation stages as well as ongoing maintenance.
Vuk Magdelinic | CEO
+1 (416) 559-7101
Justin Hui | Sales Manager
+1 (289) 544-7975