Skip to content
Forecasting for Stationary and Non-Stationary Time Series
R C++
Branch: develop
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
R
demo
inst/examples
man
src
tests
.gitignore
.travis.yml
DESCRIPTION
NAMESPACE
NEWS
README.md

README.md

Build Status Coverage Status CRAN RStudio mirror downloads CRAN_Status_Badge

forecastSNSTS: Forecasting of Stationary and Non-Stationary Time Series

The forecastSNSTS package provides methods to compute linear h-step prediction coefficients based on localised and iterated Yule-Walker estimates and empirical mean square prediction errors from the resulting predictors.

It is intended to support the paper Predictive, finite-sample model choice for time series under stationarity and non-stationarity, which we refer to as Kley et al. (2017).

You can track (and contribute to) the development of forecastSNSTS at https://github.com/tobiaskley/forecastSNSTS. If you encounter unexpected behaviour while using forecastSNSTS, please write an email or file an issue.

Getting started with forecastSNSTS

First, if you have not done so already, install R from http://www.r-project.org (click on download R, select a location close to you, and download R for your platform). Once you have the latest version of R installed and started execute the following commands on the R shell:

install.packages("forecastSNSTS")
devtools::install_github("tobiaskley/forecastSNSTS", ref="develop")

This will first install the R package devtools and then use it to install the latest (development) version of forecastSNSTS from the GitHub repository. In case you do not have LaTeX installed on your computer you may want to use

Now that you have R and forecastSNSTS installed you can access all the functions available. To load the package and access the help files:

library(forecastSNSTS)
help("forecastSNSTS")

A demo is available. It can be started by

demo("tvARMA11")

At the bottom of the online help page to the package you will find an index to all the help files available.

Replicating the examples of the paper with forecastSNSTSexamples

Note that there is a separate R package, called forecastSNSTSexamples and available only on GitHub, that can be used to replicate the empirical examples from Section 5 of Kley et al. (2017).

You can’t perform that action at this time.