Enhanced Seasonal Decomposition of Time Series by Loess

Decompose a time series into seasonal, trend, and remainder components using an implementation of Seasonal Decomposition of Time Series by Loess (STL) that provides several enhancements over the STL method in the stats package. These enhancements include handling missing values, providing higher order (quadratic) loess smoothing with automated parameter choices, frequency component smoothing beyond the seasonal and trend components, and some basic plot methods for diagnostics.

stlplus

This package contains enhancements to the Seasonal Trend Decomposition using Loess (STL) implementation that comes with base R, stl().

Here are some of the added features over stl():

• Can handle NA values
• Higher order loess smoothing (more than just local constant and linear)
• Automated parameter choices for local quadratic
• Frequency component smoothing beyond seasonal and trend
• Plot methods for diagnostics

For (very) experimental inference, prediction, and variance reduction at endpoints, see the operator package.

References

• Cleveland, R. B., Cleveland, W. S., McRae, J. E., & Terpenning, I. (1990). STL: A seasonal-trend decomposition procedure based on loess. Journal of Official Statistics, 6(1), 3-73.
• Hafen, R. P. "Local regression models: Advancements, applications, and new methods." (2010). https://docs.lib.purdue.edu/dissertations/AAI3413793/

Installation

CRAN:

install.packages("stlplus")

Github:

devtools::install_github("hafen/stlplus")

License

This software is released under the BSD license. Please read the license document.