Summary

We have started our journey beyond baseline forecasting methods and dipped our toes into the world of machine learning. After a brief refresher on machine learning, where we looked at key concepts such as overfitting, underfitting, regularization, and so on, we saw how we can convert a time series forecasting problem into a regression problem from the machine learning world. We also developed a conceptual understanding of different embeddings, such as time delay embedding and temporal embedding, which can be used to convert a time series problem into a regression problem. To wrap things up, we also learned about a new paradigm in time series forecasting – global models – and contrasted them with local models on a conceptual level. In the next few chapters, we will start putting these concepts into practice, and see techniques for feature engineering, and strategies for global models.