7
Machine Learning-Based Approaches to Time Series Forecasting
In the previous chapter, we provided a brief introduction to time series analysis and demonstrated how to use statistical approaches (ARIMA and ETS) for time series forecasting. While those approaches are still very popular, they are somewhat dated. In this chapter, we focus on the more recent, ML-based approaches to time series forecasting.
We start by explaining different ways of validating time series models. Then, we move on to the inputs of ML models, that is, the features. We provide an overview of selected feature engineering approaches and introduce a tool for automatic feature extraction that generates hundreds or thousands of features for us.
Having covered those two topics, ...
Get Python for Finance Cookbook - Second Edition now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
