View source: R/coint.test.GH.R
coint.test.GH | R Documentation |
Gregory and Hansen (1996) test for the null hypothesis of no cointegration under a possible structural break at the unknown moment of time.
The authors proposed ADF- and Z-type tests, slightly modified to allow the presence of a possible regime shift. Three type of shifts are allowed:
a shift in the constant,
a shift in the constand with the trend included,
and a shift in the constant and the cointegrating vector.
Critical values are calculated via the adopted MacKinnon procedure of estimating the model for the response surface.
coint.test.GH(
...,
shift = "level",
trim = 0.15,
max.lag = 10,
criterion = "aic",
add.criticals = TRUE
)
... |
Variables of interest. |
shift |
Expected break type. |
trim |
The trimming parameter to calculate break moment bounds. |
max.lag |
The maximum number of lags for the internal ADF testing. |
criterion |
The criterion for lag selection. |
add.criticals |
Whether critical values are to be returned. This argument is needed to suppress the calculation of critical values during the precalculation of tables needed for the p-values estimating. |
An object of type cointGH
. It's a list of
shift
: shift type,
Za
: MZ_\alpha
statistic and c.v.,
Zt
: MZ_t
statistic and c.v.,
ADF
: ADF
statistic and c.v..
MacKinnon, James G. “Critical Values for Cointegration Tests.” Working Paper. Working Paper. Economics Department, Queen’s University, January 2010. https://ideas.repec.org/p/qed/wpaper/1227.html.
Gregory, Allan W., and Bruce E. Hansen. “Residual-Based Tests for Cointegration in Models with Regime Shifts.” Journal of Econometrics 70, no. 1 (January 1, 1996): 99–126. https://doi.org/10.1016/0304-4076(69)41685-7.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.