Why You (Currently) Do Not Need Deep Learning for Time Series Forecasting
What you need instead: Learnings from the Makridakis M5 competitions and the 2023 Kaggle AI report
Deep Learning for Time Series Forecasting receives a lot of attention. Many articles and scientific papers write about the latest Deep Learning model and how it is much better than any ML or statistical model. This gives the impression that Deep Learning will solve all our problems for time series forecasting. In particular for new people in the field.
But from my experience, Deep Learning is not what you need. Other things are more important and work better for time series forecasting.
Hence, in this article, I want to show you what works. I will show you things that have proven themselves in many ways. I will use the findings of the Makridakis M5 competitions and the Kaggle AI Report 2023 and compare them against my experience.
The Makridakis competitions compare forecasting methods on real-world data sets. They show what works in practice. Since the start of the competitions almost 40 years ago, their findings have changed. In the first three competitions, M1 to M3, statistical models dominated the field. In the M4 competition, ML models began to show their potential in the form…
