Vassilios Assimakopoulos

TL;DR: It is found that the post-sample accuracy of popular ML methods are dominated across both accuracy measures used and for all forecasting horizons examined, and that their computational requirements are considerably greater than those of statistical methods.

TL;DR: All aspects of M4 are covered in detail, including its organization and running, the presentation of its results, the top-performing methods overall and by categories, its major findings and their implications, and the computational requirements of the various methods.

TL;DR: The M4 competition is the continuation of three previous competitions started more than 45 years ago whose purpose was to learn how to improve forecasting accuracy, and how such learning can be applied to advance the theory and practice of forecasting.

TL;DR: It is found that forecasting accuracy is influenced as follows: for fast-moving data, cycle and randomness have the biggest (negative) effect and the longer the forecasting horizon, the more accuracy decreases, and for intermittent data, inter-demand interval has bigger impact than the coefficient of variation.

TL;DR: A non-systematic review of the theory and the practice of forecasting, offering a wide range of theoretical, state-of-the-art models, methods, principles, and approaches to prepare, produce, organise, and evaluate forecasts.