65
at a 5% significant level and also the data are integrated of order one 1(1), as
the Augmented Dickey and Fuller (ADF) test was employed to carry out the
test. The AIC criterion was used to determine the optimal Lag length also.
Table 5: ADF- Unit Root Test for Stationarity
Method
Statistic
Prob.**
ADF - Fisher
Chi-square
49.8811
0.0000
ADF - Choi
Z-stat
-5.73867
0.0000
** Probabilities for Fisher tests are computed using an asymptotic Chi
-square distribution. All other tests assume asymptotic normality.
Intermediate ADF test results D(UNTITLED)
Series
Prob.
Lag
Max Lag
Obs
D(ROA)
0.0064
0
2
18
D(LOAN_VOLUME) 0.0044
1
2
17
D(INT_RATE)
0.0006
0
2
18
D(DEBT)
0.0009
0
2
18
Source: EViews 9 computation of reseach data.
5.4 Test for Multicollinearity Results
The multicollinearity test was conducted in order to determine the relationship
between the variables and the result are as follows:
Table 6: Test for Multicollinearity:
Correlation Matrix
ROA
LOAN_VOLUME INT_RATE
DEBT
ROA
1.000000
0.891139
0.765346
-0.540509
LOAN_VOLUME 0.891139
1.000000
0.756027
-0.659195
INT_RATE
0.765346
0.756027
1.000000
-0.712014
DEBT
-0.540509
-0.659195
-0.712014
1.000000
Source: Eviews 9 Computation of Research Data
The table shows that there is a strong positive correlation between ROA and
Loan and Advances volume as the value recorded is 0.891139 and also a
strong negative correlation between Loans and
Advances to Debt ratio at a
value of -0.659195 and also a negative correlation between interest and debt.
The table also shows that, the Banking Sector influences the performance of
the communication sector since the variables are all below 0.8 which shall
66
make it easy to extract the coefficient estimates with a small standard error.
The data model does not show evidence of multicollinearity as all the variables
are below 0.8. Further test for multi collinearity using Variance Inflation Factor.
Do'stlaringiz bilan baham: