Artificial Intelligence in Finance
This year marked the 10th annual Swissquote Conference. Current research and insights on AI in finance provided by leading experts and scholars in the field were featured on the conference program. It was also an opportunity to inaugurate the new Finance and Technology Research Programme.
On November 8th, more than 230 people gathered together at the SwissTech Convention Center to attend the Swissquote 2019 Conference. Organized each year by the Swissquote Chair in Quantitative Finance of EPFL, this 10th edition addressed the theme of Artificial Intelligence (AI) in Finance.
AI has become a prominent tool in finance. “Formally, AI has been used in finance for more than twenty years, but it has really become prominent only in the last five years,”says Damir Filipovic, head of the Swissquote Chair in Quantitative Finance. For example, in the banking sector, AI is commonly used in office work for automatization of processes such as fraud detection or abuse/misuse of credit cards. But since the development of computer and storage power and the increasing amount of data available, the possibilities of using AI have increased considerably. “It really is a breakthrough in the sense that we can now handle problems that were not processable ten years ago in terms of complexity. It happened in other areas like computer vision and natural language processing, and is now also occurring in finance,” explains Filipovic.
Applications in all areas of finance
Applications of AI range from trading, investing, and risk management, to advisory and marketing of financial services. During the conference, leading experts and scholars in the field presented some concrete applications that they use.
For example, some algorithms trained with AI can help predict the evolution of prices in financial markets. One of the difficulties is that the economic mechanisms that generate prices, and therefore returns to investors, may change through time, so that historical data from an earlier time may tell us little or nothing about future prices and returns. To address this problem, one of the speakers, Jan Witte from University College London, presented how he uses machine-learning methods to generate “synthetic” financial data. Those data are then used to test new investment strategies and help investors plan for retirement and other personal investment goals with more realistic future return scenarios.
Various, more specific cases of trading were also presented like the optimization of hedging an options book with reinforcement learning or a short-term price forecast for optimizing trade execution in future and interest-rate markets. “These are advances that will benefit everyone. We all have pensions and a 2nd pillar that are placed by pension insurances. Hopefully, they will increase and make a profit for the economy. From now on, AI can help with that,” adds Filipovic.
Inauguration of the new Finance and Technology Research Programme
The Swissquote Conference was also the occasion to present and inaugurate the new Finance and Technology Research Programme of EPFL, known informally as Fintech Lab. Combining the specialist knowledge of the Swiss Finance Institute @ EPFL with insights from the School’s data scientists and digital trust experts, this new program will be joined by two or three PhD students and one or two postdoctoral researchers in its inaugural year. “The researchers will explore and develop new technologies with applications in fields like asset management, automated trading, risk management and corporate governance,” says Filipovic. “There will also be a strong emphasis on building trust.”