Forecasting time series with multiple seasonal patterns using Prophet

You were introduced to Prophet in Chapter 11, Additional Statistical Modeling Techniques for Time Series, under the Forecasting time series data using Facebook Prophet recipe, using the milk production dataset. The milk production dataset is a monthly dataset with a trend and one seasonal pattern.

The goal of this recipe is to show how you can use Prophet to solve a more complex dataset with multiple seasonal patterns. Preparing time series data for supervised learning is an important step, as shown in Chapter 12, Forecasting Using Supervised Machine Learning, and Chapter 13, Deep Learning for Time Series Forecasting. Similarly, the feature engineering technique for creating additional features is an important topic when training and improving machine learning and deep learning models. One benefit of using algorithms such as the UCM (introduced in the previous recipe) or Prophet is that they require little to...