Exclusive Interview | Blockchain Expert Tomaso Aste
Tomaso Aste is professor of Complexity Science at UCL Computer Science Department. A trained Physicist, has substantially contributed to research in complex structures analysis, financial systems modelling and machine learning. He is passionate in the investigation of the effect of technologies on society and currently he focuses on the application of Blockchain Technologies to domains beyond digital currencies. He is Scientific Director and Founder of the UCL Centre for Blockchain Technologies; Head and Founder of the Financial Computing and Analytics Group; Programme Director of the MSc in Financial Risk Management; Vice-Director of the Centre for doctoral Training in Financial Computing & Analytics; Member of the Board of the ESRC LSE-UCL Systemic Risk Centre. Prior to UCL he held positions in UK and Australia. He is advising and consulting for financial institutions, banks and digital-economy companies and startups.
Grace: What are your thoughts on factor-based investing and where do you see it in 10 years?
Tomaso: I can see that factor-based investing is a trending fashion. The general problem in finance and investing is that there are a lot of variables, many assets and endless sources of information and noise. It is a very complex multivariate system that evolves in time changing continuously. Factor-based investing is one of the possible ways to reduce dimensionality of the problem. Does it work? Yes, to some extent if applied with the right care. Is it the only way? Certainly not, there are several ways to reduce dimensionality. I believe the most promising avenue concerns the combination of network theory and parsimonious modelling to reduce dimensionality but keep the ability to model the complexity of markets effectively.
I believe the most promising avenue [of factor-based investing] concerns the combination of network theory and parsimonious modelling
Grace: How can the Hurst exponent be used to evaluate the level of stability/instability in financial markets?
Tomaso: The Hurst exponent is a tool to quantify complexity in the time-scales domain. It is still a non-fully explored perspective that has a lot of potential. Markets behave differently at different time-horizons, Hurst exponent and other measures such as the Empirical Mode Decomposition can quantify how market dynamics changes with time-horizons. Of course, unstable periods have larger changes but also are less-random and this can be observed with the Hurst exponent method.
Grace: You did research that showed the consistently decreasing centrality of the financial sector over the last 10 years. Can you tell us more about this and if you expect this trend to continue in the future?
Tomaso: This was one of the outcome of one of our researches about networked structure of financial markets. Yes finance is becoming something more complex by itself mixing with other sectors in a non-simple way. Loss of centrality does not mean loss of importance. Traditional finance will slowly become marginal, but new finance is booming.
Grace: What are your thoughts on cryptocurrencies being used by central banks to maintain control over the monetary supply?
Tomaso: I believe that cryptocurrencies will take over, but they will still be– at the end– mostly under the control of states or large institutions. However, these institutions might not be the present banks.
Grace: The majority of trading volume (approx. 60%) is being done by quant investing. How do you see this affecting volatility in the market?
Tomaso: Quant investing is moving markets and of course strongly impacting volatility. What I believe is worrying is the fact that the industry uses very similar models and algorithms and this lack of heterogeneity makes the entire ecosystem fragile. Feedback loops and self-reinforcements can have massive effects.
Quant investing is moving markets and strongly impacting volatility
Grace: You did research on social media being used in the prediction of markets. Do you think that the sentiment of the social media messages or the volume of the messages social media messages is more statistically-significant in market prediction?
Tomaso: We have demonstrated that in some instances social media information is more powerful than prices for prediction. This is hardly surprising because there are emotional and social components in investment choices. Of course all these signals should be considered in market prediction. As usual, the easy catches for arbitrage are disappearing fast and in a few years it will be harder to make prediction from these signals.
Grace: How can a network approach be used to reduce investment risk and help build well diversified portfolios?
Tomaso: Risk can be reduced only if we gain some understanding of the markets and their future evolution. Networks are the main instruments to describe and model complex systems. Markets are very complex and networks are essential tools for their understanding.
Grace: What do think are some of the best ways to use machine learning?
Tomaso: Machine learning is becoming powerful. This is mainly due to the largely increased amount of available data that can now be used to train machines and make them learn from the data. There is however a very important difference between traditional Machine Learning domains, such as image recognition, and finance. In finance often we have a scarce amount of data that are good for training. Markets change rapidly, and historic behaviours are significant but not fully representative of future ones. Present machine learning tools must be adapted and used with carefulness and rigour otherwise results will not be reliable. There is a lot of research to be done in this domain.
Grace: What do you see on the horizon for the future of machine learning?
Tomaso: In my perspective the combination of machine learning with network approaches is the way to go. Potentials concern the capability to follow simultaneously a very large number of variables and use this for forecasting purposes.
Risk Warning: The FXCM Group does not guarantee accuracy and will not accept liability for any loss or damage which arise directly or indirectly from use of or reliance on information contained within the webinars. The FXCM Group may provide general commentary which is not intended as investment advice and must not be construed as such. FX/CFD trading carries a risk of losses in excess of your deposited funds and may not be suitable for all investors. Please ensure that you fully understand the risks involved.