Auto Regressive Distributed Lag (ARDL) for time series is a package to estimate dynamic models with lagged regressors and lagged dependent variable.
A single equation (univariate) model is estimated with the ARDL framework presented by Pesaran1999 and Pesaran2001. This version also supports automatic identification of the best model according to different selection criteria (BIC; AIC, R2 and LL). It also provides tools to visualize the cointegration (long-term) relation and to test it using the bounds test procedure.
ardl
is the core function that relies on package dynlm
to estimate the dynamic models.
auto.ardl
uses ardl
to find the best specification.
coint
presents the two sets of coefficients: long-run (LR) and short-run (SR).
bounds.test
tests the existence of a long-run relationship with I(0) or I(1) regressors.
Note that print
and summary
work as for any linear model.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.