Skip to main content

Time series decomposition chart

Microsoft Corporation
Understand the time series components using “Seasonal and Trend decomposition using Loess”

It is a common scenario: A practitioner has sales data for the past several months and wants to make sense of the time series data. Typically next step would be to perform a forecast for the next time period. The decomposition of time series is a statistical method that splits a time series into several components, each representing one of the underlying processes. There are three components that are typically of interest: trend, seasonality and noise. Time series decomposition is an essential analytics tool to understand time series components and to improve a forecast. You can control the algorithm parameters and the visual attributes to suit your needs. The current visual implements the well-known “Seasonal and Trend decomposition using Loess” approach.

Highlighted features:

  • The underlying algorithm requires the input data to be equally spaced time series
  • Seasonal factor can be found automatically or set by user
  • The choice of additive or multiplicative model can be performed automatically or set by user
  • Time series decomposition is a powerful analytics tool. Seven different modes of time series visualization are provided to allow the analyst to drill down into different aspects of data
  • Advanced user may control the inner parameters of the algorithm

R package dependencies(auto-installed): proto, zoo

Supports R versions: R 3.3.1, R 3.3.0, MRO 3.3.1, MRO 3.3.0, MRO 3.2.2

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

Add-in capabilities
When this add-in is used, it
  • Can read and make changes to your document
  • Can send data over the Internet