In the variate framework, the bounds test examines whether a long-run
relationship exists in one of the following unrestricted error correction models:In equation (1), the null hypothesis of no co-integration amongst the variables is H0:
a1a20 against the alternative hypothesis of H1: {a1 " 0}@{a2 " 0}. In equation
(2), the null hypothesis of no co-integration amongst the variables is H0: b1b20
against the alternative hypothesis of H1: {b1 " 0}@{b2 " 0}. The null hypothesis
can be tested with the F-test. But, the F-test has a non-standard distribution. Pesaran
et al. (2001) provide the critical values at table CI(iii). At k1, the critical value
bounds are (4.04, 4.78) at the 10% significance level, (4.94, 5.73) at the 5%
significance level and (6.84, 7.84) at the 1% significance level. To minimize the loss
of degree of freedom and to fulfill the assumption of no autocorrelation required by
the bounds test, the value of n corresponding to each equation is increased till the
Breusch-Godfrey Lagrange multiplier test is unable to reject the null of no
autocorrelation with lag order 2 at the 5% significance level. The results of the
bounds test are reported in Table 4.
Results in Table 4 indicate that the null of no co-integration is rejected at the 5%
level for both equations. It is clear that there is a long-run relationship between
OFDI and GDPP when either OFDI or GDPP is assigned as the dependent variable.
To obtain the long-run coefficients, ARDL models shown are estimated:
In the variate framework, the bounds test examines whether a long-runrelationship exists in one of the following unrestricted error correction models:In equation (1), the null hypothesis of no co-integration amongst the variables is H0:a1a20 against the alternative hypothesis of H1: {a1 " 0}@{a2 " 0}. In equation(2), the null hypothesis of no co-integration amongst the variables is H0: b1b20against the alternative hypothesis of H1: {b1 " 0}@{b2 " 0}. The null hypothesiscan be tested with the F-test. But, the F-test has a non-standard distribution. Pesaranet al. (2001) provide the critical values at table CI(iii). At k1, the critical valuebounds are (4.04, 4.78) at the 10% significance level, (4.94, 5.73) at the 5%significance level and (6.84, 7.84) at the 1% significance level. To minimize the lossof degree of freedom and to fulfill the assumption of no autocorrelation required bythe bounds test, the value of n corresponding to each equation is increased till theBreusch-Godfrey Lagrange multiplier test is unable to reject the null of noautocorrelation with lag order 2 at the 5% significance level. The results of thebounds test are reported in Table 4.Results in Table 4 indicate that the null of no co-integration is rejected at the 5%level for both equations. It is clear that there is a long-run relationship betweenOFDI and GDPP when either OFDI or GDPP is assigned as the dependent variable.To obtain the long-run coefficients, ARDL models shown are estimated:
การแปล กรุณารอสักครู่..