Corvil Taps AI For ’12 Steps of Big Data Grief’
Corvil, which provides data analytics for electronic trading, is using machine learning and artificial intelligence to help clients make sense of huge volumes of data.
David Murray, chief business development officer at Corvil, told Markets Media that market participants are looking for help in using data in three areas – client intelligence, measuring execution quality, and venue analysis. They find it difficult to analyse huge amounts of data and determine which signals are most important.
Murray told Markets Media: “Clients are struggling with what we call the ’12 steps of Big Data grief’ while Corvil has rich data covering the order lifecycle. The Intelligence Hub has been developed over three years to provide analytics that can be used easily in electronic trading.”
He said the quality, precision and normalization of data is critical for using machine learning successfully. Corvil captures trade details such as the precise time of a transaction, how long it took, the counterparty, and then normalizes and enriches the information.
The firm has launched Intelligence Hub, which uses machine learning to provide digital intelligence and sends proactive alerts in real-time on signals which are most important to specific clients. Murray continued that clients can choose the data points they want to monitor, such as the ratio of orders to executions.
Last year Corvil launched ‘Cara’ which uses machine learning algorithms to safeguard electronic trading businesses without affecting performance which the firm said was the first virtual security expert to address cyber security in financial markets.
Murray added: “Clients need to ingest huge volumes of data and we have created several machine learning algos to correlate anomalies. This is a self-service solution for the business, operations and compliance.”
Intelligence Hub provides easy-to-use, visualizations so users can perform multidimensional analysis. “For example, you may want to look at at venues which provide the greatest success with specific symbols or counterparties with whom you have the highest hit rate,” Murray added.
Brad Bailey, research director, capital markets at consultancy Celent, said in a statement: “Given the massive trade and order messaging rates that keep growing both in volume, speed and across asset classes, capital market players need to better understand the state of their orders/fills/data with nanosecond precision. The cost of downtime, venue or counterparty issues, and the associated risks are so great, trading and operations need to understand, analyse, and predict these immediately. These data sets are fertile ground to leverage machine learning to the utmost in the capital markets.”
Fergal Toomey, chief scientist and co-founder said on Corvil’s blog that artificial intelligence will become an invaluable assistant in troubleshooting. For example, by sifting through network data and helping resolve anomalies more quickly. Artificial intelligence can also help firms meet the requirements of MiFID II, the regulations that went live in the European Union this year.
“MiFID II mandates that financial traders have to stay within designated ratio limits, but it can be difficult in the course of a busy day to know how close they are to hitting the limit,” Toomey added. “Based on past experience, machine learning can predict where they are likely to finish up at and alerts can be triggered to prompt traders to take corrective action.”
Corvil has predominantly been used by banks but Murray expects that the buy side will be a key growth area as asset managers increasingly want to evaluate the performance of their brokers.
Mediocre venue analysis tools
Buy-side firms in the US rated their tools for venue analysis as mediocre according to a survey by consultancy Greenwich Associates, which was commissioned by Clearpool Group.
The report, Venue Analytics: Routing a Path to Best Execution, said traders need better tools to help them understand the performance of each venue and the impact of routing behavior.
One trader at an institutional asset manager said in the report: “Venue analysis will become more important to regulators and clients as we transition into MiFID II world.”
Richard Johnson, vice president in the market structure and technology practice at Greenwich said in the report that the use of disparate tools and tepid satisfaction levels for venue analysis do not suggest that the practice lacks usefulness but only that current implementations remain in the early innings.
“This represents an opportunity for vendors to develop robust, intuitive tools that will enable all traders to perform sophisticated venue analysis without needing to invest significant time and resources into the process,” added Johnson. “With coming regulatory actions sure to increase focus on routing and asset owners already taking it into consideration, venue analysis is set to become an even more important tool for ensuring best execution.”
NeoXam helps asset managers consolidate data from different providers.
Proper governance is not a roadblock to development.
SynFinance uses AI to identify investors most likely to participate in a syndicated loan.
Qplum turns to HFT data to meet its data needs.
Success relies on buy in from the rank and file, not just the IT department.