t] is estimated for each country and the CRDW
is assigned to the end date of the estimation interval.
To sum up, our conclusion is based on both the AEG and CRDW
tests giving the variables that DRPPI, gold price index, total direct foreign trade and number of completed residential units are cointegrated.
For critical values of CRDW
, see Engle and Granger (1991), Table 2 and for ADF, see MacKinnon (1991).
In this section, we have used the Engle-Granger test and CRDW
test [see Sargan and Bhargava (1983)] to investigate whether the variables under question are cointegrated or not.
The 5% critical values for the CRDW
, CR(A)DF and [Mathematical Expression Omitted] tests are, respectively: 1.
statistics reject the presence of unit roots in all cases.
Two test statistics are used in conjunction with each cointegration test: while ADF and CRDW
statistics are used with the Engle-Granger test, [[?
The book presents ADF, DF, and CRDW
statistics to test for non-stationarity and order of cointegration.
A cursory look at the table is sufficient to see that the DF, ADF and CRDW
(10) statistics reject the null hypothesis of no cointegration at the 5 percent level of significance for all the three equations.
For the residuals to be stationary, the CRDW
must be significantly different from zero.
Moving to Table 3, and concentrating on rows 18 and 20, we notice that the value of CRDW
implies that the AR(1) in [[epsilon].
Alternative to DF/ADF tests, the CRDW
statistic could also be used to determine the stationarity of residuals.