Data science in economics and finance: tools, infrastructure and challenges

Bruno Tissot

Interest in “data science” has expanded in parallel with the surge of data generated by human activities since the start of the 2000s. This concept basically relates to the application of mathematical tools to analyse data; from this perspective, it is not very different from statistics, which encompasses the various quantitative methods used for collecting, organising, analysing, interpreting and presenting data. Yet data science is usually intrinsically understood as being applied to very large and complex data sets, generally described as big data; from this perspective, it combines the application of traditional statistical methods with state-of-the-art IT techniques to deal with the vast amount of information that cannot be exploited by a single human. In other words, data science requires the support of computing machines to run long or complex mathematical calculations, and basically represents the mix of statistical techniques, advanced mathematical calculations and IT operations needed to deal with big data effectively.

Just as data science can be a multi-faceted concept, big data is also not so easy to define precisely; in general, it refers to the proliferation of