dynardl.auto.correlated: Run a variety of autocorrelation tests on the residuals from...

In andyphilips/dynamac: Dynamic Simulation and Testing for Single-Equation ARDL Models

Description

Run a variety of autocorrelation tests on the residuals from a dynardl model

Usage

dynardl.auto.correlated(x, bg.type = "Chisq", digits = 3,
  order = NULL, object.out = FALSE)

Arguments

x	a dynardl model
bg.type	a character string for the type of Breusch-Godfrey test to run. The default is Chisq: the Chisq test statistic. The other option is F: the F-test statistic
digits	the number of digits to round to when showing output. The default is 3
order	the maximum order of serial autocorrelation to test when executing the Breusch-Godfrey test
object.out	if TRUE, and dynardl.auto.correlated is assigned to an object, the AIC, BIC, and results will be stored for the user's convenience

Details

This is a simple and convenient way to test whether the residuals from the dynardl model are white noise. As an aside, this is also why dynardl has a simulate = FALSE argument: users can ensure the model has white noise residuals before estimating a potentially time-intensive simulation. The output also reminds the user of the null hypotheses for the autocorrelation tests

Value

The results of autocorrelation tests

Author(s)

Soren Jordan and Andrew Q. Philips

Examples

# Using the ineq data from dynamac
ardl.model <- dynardl(concern ~ incshare10 + urate, data = ineq, 
       lags = list("concern" = 1, "incshare10" = 1),
       diffs = c("incshare10", "urate"), 
       lagdiffs = list("concern" = 1),
       ec = TRUE, simulate = FALSE)
dynardl.auto.correlated(ardl.model)

andyphilips/dynamac documentation built on Oct. 14, 2020, 9:48 p.m.