Data is considered the oil of the 21st century, undoubtedly due to the value of the insight associated with it. If that is true, then the data processors are modern-day refineries taking a crude product and turning it into a valuable asset. Unlike oil, which is a limited resource, data is limitless and ever-expanding. Being able to process massive amounts of data requires incredible amounts of computing power and the programming prowess to make sense of it all.
Artificial intelligence (AI) is a wide-ranging branch of computer science concerned with building smart machines capable of performing tasks that typically require human intelligence. AI is an interdisciplinary science with multiple approaches, but advancements in Machine Learning (ML) and deep learning are creating a paradigm shift in virtually every industry.
Imandra is the world-leader in cloud-scale automated reasoning, the Imandra automated reasoning engine is built on recent advances in AI. It provides a scalable solution for extracting actionable intelligence from data produced by the most complex trading systems. Imandra’s partnership with Itiviti is leveraging this technology to change how financial firms connect, test and certify their systems. Trends are showing firms are becoming much more open (if not demanding) to be able to easily switch technology partners more frequently as their business models evolve.
"Efficient and rigorous monitoring, auditing, client onboarding and certification require a thorough analysis of data... Machine Learning is the go-to tool for this job."
Financial markets run on myriad complex systems communicating with each other. They implement various protocols, must abide by numerous rules and regulations, and in the process, generate a significant amount of data. Efficient and rigorous monitoring, auditing, client onboarding and certification require a thorough analysis of this data. Historically, much of this analysis has been manual and thus error-prone and expensive. Machine Learning is the go-to tool for this job.
ML is used as an all-encompassing phrase and finance is no stranger to it. In fact, it's difficult to find a hedge fund, bank or a trading platform today that does not leverage some form of AI/ML in its day-to-day work. Those use cases, however, are fundamentally different from the analysis of complex inter-system communication data (that uses FIX, for example). Traditional ML approaches are used to forecast asset prices, market volatility or detect market abuse/fraud. In all of those cases, the data may be normalized so that a neural network or a decision tree can be trained on it.
Transactional data, for example, is exchanged between a buy-side’s trading system and an investment bank, contains massive amounts of counterparty logic (e.g. algo strategy type and its parameters, enrichment rules, customized client-centric logic, etc). This tag mapping and behavior have traditionally not been well documented and as a result, creates angst to the sell-side when entertaining changing connectivity/platform providers. This data simply cannot be shoe-horned for analysis by traditional ML techniques used in classification/forecasting.
What's required is the ability to extract complex logical patterns from that data (the technical term is 'rule synthesis'), combine them into a single logical model of the interaction and be able to leverage the resulting model for audit, testing, etc. (see diagram). This was not possible only a few years ago - the science simply wasn't there. The partnership between Imandra and Itiviti will bring these deep advances in AI for algorithms to financial infrastructure on a massive scale.
"The partnership between Imandra and Itiviti will bring these deep advances in AI for algorithms to financial infrastructure on a massive scale."
The same advances have allowed us to scale AI to our financial testing software. Imandra will help us automate the analysis of a FIX specification (or trading algorithms for instance) and generate all of the unique test cases (including the edge cases) required fully describing the interface logic. This testing logic allows brokers to ensure that they can fully certify connections with new counterparties as well as properly back-test their logic changes. Imandra's test generation is seamlessly integrated into Itiviti's automated onboarding tools testing products and is poised to bring much-needed efficiency and rigor to the testing space.
Download the full article
Want to learn more?
Data Strategy - It's Morphin’ time
As I watched my third episode of Power Rangers with my daughter where the world is again under...
What to consider when changing OMS / EMS
The order management system (OMS) and execution management system (EMS) marketplace is one where historically firms have shied...
Deriving value from managed FIX services BI analytics
Itiviti, in conjunction with GreySpark Partners, recently engaged in a comprehensive assessment of how brokers use vendor-provided managed...
George is Head of Product Strategy- Client Connectivity and Data Analytics at Itiviti. Prior to working at Itiviti, George worked as a Managing Director at Convergex, an Agency Broker Dealer, where he was responsible for the Trading Services Division which included Global Trade Support, Global Connectivity, Vendor/Exchange/Broker Relations as well as the CONNEX Managed Services business, which he founded while there. George holds a Bachelor of Science in Business Management from Towson University.