How smart is your algorithm?

Big data and smart algorithms that can predict what you are going to do before you even know it yourself… We have all heard these success stories about the world of big data analytics. But are big data and smart algorithms really the solution to all problems?

Stephan Smeekes’s main research interests lie in the statistical analysis of time series data, combining techniques on the interface between econometrics, statistics and data science. Much of his research involves uncertainty quantification, often using the bootstrap. Among the applications he considers are the analysis of high-dimensional (big data) time series, long-run trends in macroeconomic and climatological time series, inference on risk measures for volatile financial series, and the forecasting of economic and financial time series.

Etienne Wijler is currently a post-doctoral researcher in econometrics with a specialisation in high-dimensional time series methods. The analysis of high-dimensional datasets, popularized under the name Big Data, is becoming increasingly important for a wide range of institutions. For example, governmental agencies can benefit from improved inference on large micro-economic datasets without the need for aggregation to a macro-level, while commercial corporations often possess overwhelming amounts of potentially useful but highly unstructured data. In recognition of the opportunities provided by non-traditional high-dimensional methods, Etienne explores the behaviour of the most recent statistics from both a theoretical (asymptotic) perspective and in applied settings.