tidyverts/fable

What it solves

It provides a standardized way to perform univariate and multivariate time series forecasting. It simplifies the process of creating, evaluating, and visualizing forecasts using a consistent workflow for data scientists and statisticians.

How it works

The package implements a variety of forecasting models, including automatic ARIMA modelling and exponential smoothing via state space models (ETS). It integrates with the tidyverse ecosystem in R, allowing users to pipe data into models and generate forecasts over a specified horizon.

Who it’s for

Data analysts and researchers using the R programming language who need to perform time series analysis and forecasting.

Highlights

• Support for both univariate and multivariate time series forecasting.
• Includes automatic ARIMA and ETS models.
• Consistent workflow for evaluating and combining models.
• Seamless integration with the tidyverse for data manipulation and visualization.