Machine learning and deep learning methods are often reported to be the key solution to all predictive modeling problems. An important recent study evaluated and compared the performance of many classical and modern machine learning and deep learning methods on a large and diverse set of more than 1,000 univariate time series forecasting problems. The results of this study suggest that simple classical methods, such as linear methods and exponential smoothing, outperform complex and sophisticated methods, such as decision trees, Multilayer Perceptrons (MLP), and Long Short-Term Memory (LSTM) network models. These findings highlight the requirement to both evaluate classical methods and use their results as a baseline when evaluating any machine learning and deep learning methods for time series forecasting in order demonstrate that their added complexity is adding skill to the forecast. In this post, you will discover the important findings of this recent study evaluating and comparing the performance of a classical and modern machine learning methods on a large and diverse set of time series forecasting datasets.

Artificial Intelligence (AI) has gained considerable prominence over the last decade fueled by a number of high profile applications in Autonomous Vehicles (AV), intelligent robots, image and speech recognition, automatic translations, medical and law usage as well as beating champions in games like chess, Jeopardy, GO and poker [1]. The successes of AI are based on the utilization of algorithms capable of learning by trial and error and improving their performance over time, not just by step-by-step coding instructions based on logic, if-then rules and decision trees, which is the sphere of traditional programming. In light of the above, AI found applications in the field of forecasting and a considerable amount of research has been conducted on how a special class of it, utilizing Machine Learning methods (ML) and especially Neural Networks (NNs), can be exploited to improve time series predictions. Literally hundreds of papers propose new ML algorithms, suggesting methodological advances and accuracy improvements [2–8]. Yet, limited objective evidence is available regarding their relative performance as a standard forecasting tool [9–12].