Predictive analytics, machine learning and sentiment analysis are making their way into the trading environment, but how useful are they and will they provide a new model for intelligent trading? These questions and more will be discussed at next week’s A-Team Group Intelligent Trading Summit in New York.
Adrian Sharp, senior industry consultant, capital markets, Teradata, will moderate a panel session that will consider the potential of these emerging capabilities during the summit. He will be joined on the panel by Li Yang, vice president, lead of development, Citi; Philippe Burke, managing partner, Apache Capital; Antonio Hallak, CEO, Sibyl Trading; Steven Cohen, CEO, Gold Coast Advisors; and Yadu Kalia, worldwide business architect, financial services, IBM.
Moving on from high frequency algo trading, which is driven by speed but ultimately limited by system power, the panel will consider how the trading model may change over time to become less about speed and more about predictive analytics that could, perhaps, identify what might happen in the next 10 milliseconds.
Machine learning and sentiment analysis, which introduces unstructured data, also reduce speed. If these technologies are deployed in trading, a use case that includes a different view of time to that used in high frequency trading is required.
Sharp suggests there is a place in trading for machine learning, sentiment analysis and predictive analytics, but says it is difficult to find use cases for the technologies at the moment. He explains: “If machine learning, sentiment analysis and predictive analytics are applied to trading, the trading process changes and the model is different to that used for high frequency trading. The question is whether we are going to continue to exploit inefficiencies in the market or take a broader view over a longer time by bringing in more analytics.”
To find out more about:
- Alternative trading models
- The power of predictive analytics
- Use cases for machine learning
- Future trading developments