Summary
Hopefully, you experienced no issues in installing Prophet on your machine at the beginning of this chapter. The potential challenge of installing the Stan dependency is greatly eased by using the Anaconda distribution of Python. After installation, we looked at the CO2 levels measured in the atmosphere 2 miles above the Pacific Ocean at Mauna Loa in Hawaii. We built our first Prophet model and, in just 12 lines of code, were able to forecast the next 10 years of CO2 levels.
After that, we inspected the forecast DataFrame and saw the rich results that Prophet outputs. Finally, we plotted the components of the forecast – the trend, yearly seasonality, and weekly seasonality – to better understand the data’s behavior.
There is a lot more to Prophet than just this simple example, though. In the next chapter, we’ll take a deep dive into the equations behind Prophet’s model to understand how it works.
