Learn R Programming

tsDyn

Package tsDyn implements a variety of non-linear time series models. To read more about it, look at:

the wiki
the FAQ

Installation

You can install the released version of tsDyn from CRAN with:

install.packages("tsDyn")

Development version

You can download the version from github using:

library(remotes)
remotes::install_github("MatthieuStigler/tsDyn")

Copy Link

Version

Install

install.packages('tsDyn')

Monthly Downloads

699

Version

11.0.6

License

GPL (>= 2)

Issues

Pull Requests

Stars

Forks

Repository

https://github.com/MatthieuStigler/tsDyn/wiki

Maintainer

Matthieu Stigler

Last Published

November 3rd, 2026

Functions in tsDyn (11.0.6)

Estimation of Vector error correction model (VECM)

Forecasting accuracy measures.

Additive nonlinear autoregressive model

Simulation and bootstrap a VECM or bivariate TVECM

No cointegration vs threshold cointegration test

Test of linear cointegration vs threshold cointegration

Threshold Vector Error Correction model (VECM)

Long-term mean of an AR(p) process

Extract cointegration parameters A, B and PI

availableModels

Available models

computeGradient

computeGradient

Linear AutoRegressive models

Multivariate linear models: VAR and VECM

Trivariate time series plots

Bivariate time series plots

Time series of PPI used as example in Bierens and Martins (2010)

autotriples.rgl

Interactive trivariate time series plots

Virtual VECM model

Characteristic roots of the AR coefficients

Extract threshold(s) coefficient

Impulse response function

Mean Square Error

Neural Network nonlinear autoregressive model

delta test of linearity

delta test of conditional independence

Extract Log-Likelihood

Locally linear model

Monthly US unemployment

Logistic Smooth Transition AutoRegressive model

Predict method for objects of class ‘nlar’.

Plot the Error Correct Term (ECT) response

Plotting methods for SETAR and LSTAR subclasses

Predict method for objects of class ‘VAR’, ‘VECM’ or ‘TVAR’

Test of the cointegrating rank

Objects exported from other packages

sigmoid functions

setarTest_IIPUs_results

Results from the setarTest, applied on Hansen (1999) data

Forecast Error Variance Decomposition

Simulation and bootstrap of Threshold Autoregressive model (SETAR)

Extract a variable showing the regime

Residual variance

Selection of the lag with Information criterion.

Resampling schemes

NLAR common structure

Non-linear time series model, base class definition

selectHyperParms

Automatic selection of model hyper-parameters

predict_rolling

Rolling forecasts

fitted method for objects of class nlVar, i.e. VAR and VECM models.

Selection of the cointegrating rank with Information criterion.

Automatic selection of SETAR hyper-parameters

Latex representation of fitted setar models

Self Threshold Autoregressive model

Test of linearity against threshold (SETAR)

zeroyld time series

Getting started with the tsDyn package

Multivariate Threshold Vector Autoregressive model

Simulation of a multivariate Threshold Autoregressive model (TVAR)

Test of unit root against SETAR alternative with

Mean Absolute Percent Error

US monthly industrial production from Hansen (1999)

Generalized Impulse response Function (GIRF)

Test of unit root against SETAR alternative

Simulate or bootstrap a VAR model

US unemployment series used in Caner and Hansen (2001)

VAR representation

Specification of the threshold search

Test of linearity