Addressing the data sciences needs of financial services
UCD Michael Smurfit Graduate Business School’s new MSc in Financial Data Science has fintech at its core
Fintech, the merging of finance, data and technology, is transforming financial services. Photograph: iStock
A new masters programme just launched by the UCD Michael Smurfit Graduate Business School is aimed at equipping students with the market knowledge and financial data science skills to succeed in the new and emerging digital environment for banking and financial market transactions. Commencing in September, the new MSc in Financial Data Science has fintech at its core.
Fintech, the merging of finance, data and technology, is transforming financial services. New and emerging technologies such as big data analytics, blockchain, machine learning and AI, cloud computing and cryptocurrencies are transforming how business is conducted and leading to unprecedented disruptions in the banking and financial services landscape.
Prof John Cotter, chair in Quantitative Finance and an expert on fintech, explains some of the ways in which the technology is changing the face of financial markets. “The American rail company Amtrak had a crash a few years ago and its stock price fell within seconds. But there was an hour’s delay between the crash and when it was announced to the markets. Social media had reported the crash immediately though and that was picked up by robo-advisers which were advising traders to sell. That’s fintech.”
The increased use of personal online data is another example. “It turns out that our digital footprints are quite important and can be used to analyse all sorts of things about people,” says Cotter. “They can be used to look at the behaviour and conduct of people. And that’s going to be very big in the future. Banks can analyse people using their digital footprint and work out the probability of loan default. They have worked out that people with an email address like firstname.lastname@example.org were more likely to default. On a different note, people’s online information in traditional areas like occupation, age, address and so on can also be used to provide signals to assess repayment capacity.”
The technology can also be used to support investment decisions. “If you think about a person as they go through their life and career, they will make all sorts of investment decisions. That might be saving for retirement or investing directly in equity markets. One of the things they are interested in is risk and return. They really want to manage their risk, but we saw the biggest stock market movements in 10 years during 2020. There was a lot of volatility in the markets.”
The usual strategy to mitigate against volatility is diversification – spreading your investments over different asset classes, within asset classes, and across different geographies to minimise the risk of everything going south at the same time.
But that strategy is diminishing in value according to research led by Cotter. “Nowhere to Run, Nowhere to Hide: Asset Diversification in a Flat World” used machine learning technology to analyse markets over a 35-year period and found a distinct reduction in diversification potential, possibly as a result of globalisation.
That leaves investors back at square one and looking for ways to pick the best performing asset portfolio. Again, machine learning can help. While past performance is no guarantee of future performance, it can certainly offer a guide. In another piece of research led by Cotter, machine learning algorithms were applied to market performance over certain time periods and the learning from those was applied to predict performance for the subsequent time period.
Comparing those predictions with actual performance over the period enables algorithms to learn and improve in their predictive abilities. The cumulative learnings enable it to become more accurate over time.
“The results were very strong,” says Cotter.
“Stock exchanges are technology led,” he adds. “If you want to buy shares you do it on online trading accounts and the shares are procured through a system of online brokers. Ten years ago, you would have telephoned a broker and 20 years ago you would have gone to a bank. The dynamics of the market have changed. The traditional bricks-and-mortar banking model is gone as well. Everything from credit checks to payments and loans will all be done in the virtual world. We are rapidly moving to becoming a cashless society as well. The new masters programme will have a banking module and capital markets module. There will also be an incubator where students can set up their own fintech start-up.”
Students on the new programme will be able to immerse themselves in how financial markets operate and the ongoing fintech revolution. The programme includes the theory and practice of cutting-edge quantitative analysis, optimisation, data science and machine learning on real financial data sets. Graduates will also attain skills and understanding in software programming, databases and applications of machine learning in finance.
Students who opt for the incubator project will be asked to come up with their own fintech business idea and business plan which they will put into action in the NovaUCD Innovation Hub which nurtures and supports high-tech start-up companies to grow and scale globally.
The Irish financial services sector is set for strong growth, he believes. “Financial services growth is very much being driven by what’s happening in fintech. The IDA and the Department of Finance have both identified it as an area of opportunity and are talking about a potential for 10,000 new jobs in the sector. And there are 35,000 jobs in the sector here already. Ireland is ahead of the curve there and is benefiting from Brexit with people leaving the UK to study in other countries. We are also attracting students from other European countries. What we are really trying to do is support the broader financial services sector in Ireland.”