In this chapter, we introduced the applications of ML models for forecasting time series data. Before we jumped into the modeling part, we looked at the usage of the major concepts we've learned about throughout this book. We started with an exploratory analysis of the US vehicle sales series using seasonality and correlation analysis. The insights from this process allowed us to build new features, which we then used as inputs for the ML models. Furthermore, we looked at the advantages of the grid search for tuning and optimizing ML models. Last but not least, we introduced the AutoML model from the h2o package in order to complete the automation, tuning, and optimization processes for ML models.
With that, I hope you have enjoyed the learning journey that we have been on throughout this book!