LAMPIRAN F. Uji Asumsi Klasik LAMPIRAN F.1 : BANK PERSERO
a. Multikolinieritas
LDR CAR
NIM NPL
ROA BOPO
LDR 1,000000
0,371193 -0,455277 -0,866903 0,695668 -0,747998 CAR
0,371193 1,000000 -0,324699 -0,227274 0,471041 -0,286045
NIM -0,455277 -0,324699 1,000000
0,291952 -0,326576 0,451549 NPL
-0,866903 -0,227274 0,291952 1,000000 -0,745421 0,649386
ROA 0,695668
0,471041 -0,326576 -0,745421 1,000000 -0,460093 BOPO
-0,747998 -0,286045 0,451549 0,649386 -0,460093 1,000000
b. Linieritas
Ramsey RESET Test Equation: PERSERO
Specification: LDR C CAR NIM NPL ROA BOPO Omitted Variables: Squares of fitted values
Value df
Probability t-statistic
1.002897 24
0.3259 F-statistic
1.005803 1, 24
0.3259 Likelihood ratio
1.272676 1
0.2593 F-test summary:
Sum of Sq. df
Mean Squares Test SSR
7.619370 1
7.619370 Restricted SSR
189.4293 25
7.577172 Unrestricted SSR
181.8099 24
7.575413 Unrestricted SSR
181.8099 24
7.575413 LR test summary:
Value df
Restricted LogL -72.04254
25 Unrestricted LogL
-71.40620 24
c. Autokorelasi
Breusch-Godfrey Serial Correlation LM Test: F-statistic
6.081554 Prob. F2,23 0.0076
ObsR-squared 10.72307 Prob. Chi-Square2
0.0047
Test Equation: Dependent Variable: RESID
Method: Least Squares Date: 041216 Time: 05:57
Sample: 2008Q1 2015Q3 Included observations: 31
Presample missing value lagged residuals set to zero. Variable
Coefficient Std. Error
t-Statistic Prob.
C 0.663077
10.23183 0.064805
0.9489 CAR
0.304033 0.313316
0.970373 0.3420
NIM -1.236510
1.366484 -0.904884
0.3749 NPL
-0.537481 0.706228
-0.761060 0.4544
ROA -0.871348
1.696112 -0.513733
0.6123 BOPO
0.073529 0.056087
1.310977 0.2028
RESID-1 0.638675
0.196656 3.247674
0.0035 RESID-2
-0.586482 0.238334
-2.460761 0.0218
R-squared 0.345905 Mean dependent var
-1.29E-15 Adjusted R-squared
0.146833 S.D. dependent var 2.512829
S.E. of regression 2.321026 Akaike info criterion
4.739531 Sum squared resid
123.9047 Schwarz criterion 5.109593
Log likelihood -65.46274 Hannan-Quinn criter.
4.860162 F-statistic
1.737587 Durbin-Watson stat 1.883854
ProbF-statistic 0.149618
Melakukan diferensi untuk penyembuhan autokorelasi
Breusch-Godfrey Serial Correlation LM Test: F-statistic
1.236910 Prob. F2,22 0.3097
ObsR-squared 3.032407 Prob. Chi-Square2
0.2195
Test Equation: Dependent Variable: RESID
Method: Least Squares Date: 041216 Time: 06:00
Sample: 2008Q2 2015Q3 Included observations: 30
Presample missing value lagged residuals set to zero. Variable
Coefficient Std. Error
t-Statistic Prob.
C 0.011330
0.493855 0.022942
0.9819 DCAR
-0.169205 0.520982
-0.324781 0.7484
DNIM -0.369189
1.529753 -0.241339
0.8115 DNPL
-0.377158 1.665025
-0.226518 0.8229
DROA 0.990597
2.433564 0.407056
0.6879 DBOPO
0.055801 0.071109
0.784717 0.4410
RESID-1 -0.116559
0.298301 -0.390743
0.6997 RESID-2
-0.423251 0.313447
-1.350312 0.1906
R-squared 0.101080 Mean dependent var
0.000000 Adjusted R-squared
-0.184940 S.D. dependent var 2.278325
S.E. of regression 2.480069 Akaike info criterion
4.877629 Sum squared resid
135.3164 Schwarz criterion 5.251281
Log likelihood -65.16443 Hannan-Quinn criter.
4.997163 F-statistic
0.353403 Durbin-Watson stat 2.017013
ProbF-statistic 0.919364
d. Heteroskedastisitas