Hands-on Time Series Analysis with Python: From Basics to Bleeding Edge Techniques

© B V Vishwas and Ashish Patel 2020

B. V. Vishwas, A. PATELHands-on Time Series Analysis with Pythonhttps://doi.org/10.1007/978-1-4842-5992-4_8

8. Prophet

B V Vishwas¹  and Ashish Patel²

(1)

Infosys, Bengaluru, India

(2)

Cygnet Infotech Pvt Ltd, Ahmedabad, India

 

Prophet is an open source framework from Facebook used for framing and forecasting time series. It focuses on an additive model where nonlinear trends fit with daily, weekly, and yearly seasonality and additional holiday effects. Prophet is powerful at handling missing data and shifts within the trends and generally handles outliers well. It also allows you to accumulate exogenous variables to the model.

The Prophet Model

Prophet uses a decomposable time-series model.

y(t) = g(t) + s(t) + h(t