ardl: Estimate an Auto Regressive Distributed Lag (ARDL) model.
In fcbarbi/ardl: Auto Regressive Distributed Lag time series model

Description Usage Arguments Details Value References See Also Examples

Generates an Auto Regressive Distributed Lag (ARDL) model based on the number of lags of y and x.

1 2	ardl(formula, data, subset = NULL, ylag = 1, xlag = 1, case = 3, quiet = FALSE)

`formula`	Formula as in `y ~ x1 + x2 \| x3` where `x1` and `x2` may be lagged to different orders according to the vector `xlag=c(1,2)`. Note that `x3` is fixed so it is not lagged, it is generally used for dummies.
`data`	A dataframe or time referenced object with data in columns.
`subset`	(optional) Filter rows from the dataframe. Defaults to `NULL`.
`ylag`	(optional) Defaults to 1. Maximum lag of the dependent variable. Must be 1 or more.
`xlag`	(optional) Defaults to 1. Vector of the maximum lag for each of the lagging component. Must be 0 or more. Note that if omitted the regressors have lag 1.
`case`	(optional) Defaults to 3 (intercept + no trend). We use the same table of cases as Pesaran2001 the options are: `case = 1` model with no intercept, no trend; `case = 3` model with unrestricted intercept, no trend; `case = 5` model with unrestricted intercept, unrestricted trend.
`quiet`	(optional) Defaults to `FALSE`. If set to `TRUE` the routine genaerates no output, as required when called by `auto.ardl()`.

Saves an ardl object with all results to be print(), summary() or coint().

An object of class ardl.

Pesaran, M.H. and Shin, Yongcheol (1999) An Autoregressive Distributed-Lag Modelling Approach to Cointegration Analysis. Econometrics and Economic Theory in the 20th Century. Cambridge University Press.

Pesaran, M.H. and Shin, Yongcheol and Smith, Richard (2001) Bounds testing approaches to the analysis of level relationships. Journal of Applied Econometrics.

Hassler, Uwe and Wolters, Jürgen (2005) Autoregressive distributed lag models and cointegration. Discussion Papers.

auto.ardl

 
data(br_month)
m1 <- ardl( mpr~cpi+reer, data=br_month )
m2 <- ardl( mpr~cpi+reer|d_lula, data=br_month, ylag=2, xlag=c(1,1), case=5, quiet=TRUE )  
m3 <- ardl( mpr~cpi+prod+reer|d_lula, data=br_month, ylag=2, xlag=c(1,2,2), case=1 )