The impact of the banking sector development on the financial performance of the communication sector in sierra leone

Table 7: Variance Inflation factor Test for Multicollinearity 
 
Coefficient 
Uncentered 
Centered 
Variable 
Variance 
VIF 
VIF 
ROA(-1) 
0.023835 
46.99361 
2.320727 
LOAN_VOLUME 
0.067665 
91.49494 
2.982914 
INT_RATE 
0.361068 
322.8776 
4.828250 
DEBT 
0.017473 
94.46193 
3.104760 
C 
0.018559 
524.9397 
NA 
Source: EViews 9 Computation of Research Data. 
The above table shows that, no severe multicollinearity exists among the 
variables and VIF centered is less than 5. In this scenario we need to adopt 
the principle of leave the model alone. In as much as the model is safe for 
multicollinearity the test for cointegration can now be conducted.
5.5 ARDL Cointegration Test Results 
The result deduced by the unit root test shows that, the variables are stationary 
at first difference, this gives way for the test for cointegration to be conducted 
in order to verify the existence of a long-run equilibrium relationship between 
the variables. The Johansen Cointegration test was used to test, if there is one 
linear combination of the variables at least. The VAR estimation was used to 
determine the Lag length that to be used. The reports of the VAR lag order 
selection criteria indicate that, the optimal lag length, based on the AIC and SC 
is 2 lags. 
The Johansen test results which was performed to verify if a long-run 
relationship does exist between the selected variables is shown on the below 
table. A cointegration rank of one is shows by the Trace statistics, whereas the 
Max-Eigen statistics shows two cointegrating equations at a 5% significance 
level indicating that, the variables are cointegrated and there is an evidence of 
a long-run relationship.

---

67