Presenters: Ivan Svetunkov and Nikolaos Kourentzes

Abstract: Proper statistical models play an important role in forecasting providing a holistic framework for time series analysis. Because of that they can be extended to include more components and can be efficiently used in various practical decisions. Forecasting methods on the other hand, are used just to solve a specific problem and are not necessarily easily extendable. The distinction between the two is important, for example, for efficient inventory management and/or connecting forecasts with specific decisions. However, for all the analytical and decision making strengths that models introduce, they also come with some pitfalls. So, keeping in mind the difference between models and reality is essential for making correct decisions in practice. In this talk Ivan and Nikos will discuss the connection between methods, models and reality and reflect on John’s contribution to the topic.

This is a short presentation in the workshop organised in memory of John Boylan.