New column: What Data Scientists really do? We ask them for you.

The Florence Center for Data Science is launching a new format consisting in brief interviews with our D2 Seminar speakers. We are going to ask to the experts to explain what their research is about, as it was an extended version of Q&A session. Our goal is to promote a public understanding of science and to reach the readers who may be interested in knowing more about the different fields in which data are involved.


The first speaker we talked to is Leonardo Bargigli.

On the Fifth seminar D2 series that took place in the 15th of October, Leonardo Bargigli presented a study he recently conducted with G. Cifarelli, about Endogenous and Exogenous Volatility in the Foreign Exchange Market. This means the data involved in this research reveal something new about finance and the dynamics involved in trading.
Bargigli analized how high-frequency financial data vary through time, considering the behavior of market participants and the flow of market-relevant information. To do so, he used transaction data of the EUR/USD inter-dealer market in 2016, looking for volatility variables and the kind of information that are significant in market shocks.

So we asked him few questions to understand what that means, here’s what he said.

Professor Bargigli, how would you explain your results to someone who doesn’t know about heterogeneous belief and volatility but wants to understand more about financial market?

Financial markets are highly unpredictable. The reason for this unpredictability is twofold. On the one hand, markets receive information from an external world which is itself inherently complex and unpredictable. On the other hand, those who participate in the markets filter the information they receive from the external reality through their own specific vision. In short, it is the complexity of the world and the heterogeneity of economic agents that make markets unpredictable. In our model, we try to capture these two factors in the specific context of the foreign exchange market, to better understand its dynamics. To do this, we consider a model whose coefficients vary over time. This is justified precisely on the basis of complexity and heterogeneity. The results we get from the data confirm, in our opinion, that this is the right approach.

Would you give us some insight into what contribution has data analysis in your research?

Data analysis is of fundamental importance in my work, because I believe that the economy must move more and more in the wake of the natural sciences. The idea of an evidence-based economy is opposed to an old view according to which the validity of economic theory, deduced from first principles established in a normative way, did not depend on economic facts. I think this is a radically wrong path, and for this reason knowing thoroughly statistics and econometrics is indispensable for every economist.

Leonardo Bargigli is also a member of our center. More information about his research can be found in his personal page.

You can access to the recording of this and all other seminars at this link. (Registration needed)