Custom Visual Guide

Forecasting TBATS
Empty image or helper icon

Sam McKay

CEO & Founder

Forecasting TBATS
Time-series forecasting for series that exhibit multiple seasonalities using the TBATS model.

Imagine that your data is influenced by two seasonality factors. For example, an ice-cream seller expects a certain seasonality pattern that fluctuates daily, and weekly.

The TBATS model is a time-series model for series that exhibit multiple seasonalities.

You can control the visual attributes of the TBATS model to suit your needs.

Here’s how it works:

  • Define the required “Date” field (of type “date” or “date/time”)
  • Define the required “Value” field (numeric)
  • Select the required seasonalities in “Forecasting settings”
  • Use numerous formatting controls to refine the visual appearance of the plot

R package dependencies (which are auto-installed): zoo, scales, reshape2, ggplot2, plotly, forecast, lubridate, htmlwidgets, XML

Supports R versions: R 3.4.0, R 3.3.3, R 3.3.2, MRO 3.2.2

Warning: the optimization procedure for TBATS can be time consuming. It’s recommended only when your data exhibits multiple seasonalities.

NEW: support for tooltips on hover and selection.

This is an open source visual. Get the code from GitHub:

The full version of this visual (supported in Power BI Desktop only) is available from the following location:


  • Can read and make changes to your document