Lampiran 1
Uji Normalitas
One-Sample Kolmogorov-Smirnov Test
100 100 100 100 100 100
2.0547 5.2356 21.6284 88.2695 157.0646 .1600 2.8347 7.3813 20.6689 8.9831 697.3009 .3685
.240 .283 .251 .071 .492 .508
.223 .283 .236 .071 .492 .508
-.240 -.241 -.251 -.056 -.420 -.332
1.005 1.273 1.250 .711 1.222 1.079
.332 .116 .140 .692 .163 .317
N
Mean Std. Deviation Normal Parametersa,b
Absolute Positive Negative Most Extreme
Differences
Kolmogorov-Smirnov Z Asymp. Sig. (2-tailed)
ROA MSDN CAR BOPO LDR OWNER
Test distribution is Normal. a.
Calculated from data. b.
Uji Normalitas
Lampiran 2
Regression
Variables Entered/Removedb
OWNER, CAR, LDR, BOPO,
Removed Method
All requested variables entered. a.
Dependent Variable: ROA b.
Model Summaryb
.668a .447 .417 2.1643 2.092
Model 1
R R Square
Adjusted R Square
Std. Error of the Estimate
Durbin-W atson
Predictors: (Constant), OWNER, CAR, LDR, BOPO, MSDN a.
Dependent Variable: ROA b.
R sebesar 0.668, berarti bahwa hubungan antara Variabel Dependent dan Variabel Independet tergolong kuat (>0.5) R Square sebesar 0.447, berarti bahwa sebesar 44.7% perubahan yang terjadi pada variabel Dependent
dipengaruhi/disebabkan oleh variabel Independent, sedang sisanya sebesar 55.3% (100%-44.7%) dipengaruhi oleh faktor lain yang tidak dimasukkan pada model.
ANOVAb
355.206 5 71.041 15.166 .000a
440.308 94 4.684
795.514 99
Regression
Squares df Mean Square F Sig.
Predictors: (Constant), OWNER, CAR, LDR, BOPO, MSDN a.
Dependent Variable: ROA b.
Uji F
Coefficientsa
6.717 2.263 2.969 .004
8.897E-03 .044 .023 .202 .840 .448 2.233
8.286E-02 .011 .604 7.750 .000 .969 1.032
-7,33E-02 .025 -.232 -2.960 .004 .956 1.046
-3,62E-04 .000 -.089 -1.140 .257 .965 1.036
.175 .877 .023 .200 .842 .453 2.208
(Constant)
B Std. Error Unstandardized
t Sig. Tolerance VIF
Collinearity Statistics
Dependent Variable: ROA a.
Dari hasil analisis diatas didapat Model persamaam Regresi : ROA = 6.717 + 0.008897 MSDN + 0.08286 CAR - 0.0733 BOPO - 0.000362 LDR + 0.175 OWNER + e
Uji T
Pada tabel diatas variabel memiliki CAR dan BOPO memiliki t uji dengan tingkat signifikansi < 0.05, maka hal ini menandakan bahwa secara parsial variabel independent berpengaruh signifiakan terhadap variabel dependent. Dan sebaliknya pada variabel MSDN, LDR dan OWNER memiliki signifikansi t uji > 0,05. Hal ini menandaka bawa variabel bersangkutan tidak berpengaruh signifikan terhadap ROA
Coefficient Correlationsa
1.000 -.049 -.028 -.112 -.738
-.049 1.000 -.088 .125 .075
-.028 -.088 1.000 .117 .088
-.112 .125 .117 1.000 .123
-.738 .075 .088 .123 1.000
.769 -4,56E-04 -7,71E-06 -2,43E-03 -2,85E-02 -4,56E-04 1.143E-04 -2,99E-07 3.305E-05 3.523E-05 -7,71E-06 -2,99E-07 1.008E-07 9.238E-07 1.231E-06 -2,43E-03 3.305E-05 9.238E-07 6.134E-04 1.343E-04 -2,85E-02 3.523E-05 1.231E-06 1.343E-04 1.939E-03 OWNER
1 OWNER CAR LDR BOPO MSDN
Collinearity Diagnosticsa
4.734E-03 26.888 .99 .03 .03 .99 .02 .02
Dimension
Index (Constant) MSDN CAR BOPO LDR OWNER
Variance Proportions
Dependent Variable: ROA a.
Casewise Diagnosticsa
5.464 24.62
-4.673 .90
Case Number 99
100
Std. Residual ROA
Dependent Variable: ROA a.
Residuals Statisticsa
-5.1048 12.7935 2.0547 1.8942 100
-10.1134 11.8265 8.882E-16 2.1089 100
-3.780 5.669 .000 1.000 100
-4.673 5.464 .000 .974 100
Predicted Value Residual
Std. Predicted Value Std. Residual
Minimum Maximum Mean Std. Deviation N
Lampiran 3
Uji Autokorelasi
Model Summaryb
.668a .447 .417 2.1643 2.092
Model 1
R R Square
Adjusted R Square
Std. Error of the Estimate
Durbin-W atson
Predictors: (Constant), OWNER, CAR, LDR, BOPO, MSDN a.
Dependent Variable: ROA b.
Uji Asumsi Klasik Autokorelasi
Dari hasil Model Summary diatas didapat hasil d = 2.092, sedangkan dari tabel Durbin Watson dengan n = 100 dan k = 5 didapat,
Lampiran 4
Uji Multikolinieritas
Coefficientsa
6.717 2.263 2.969 .004
8.897E-03 .044 .023 .202 .840 .448 2.233
8.286E-02 .011 .604 7.750 .000 .969 1.032
-7,33E-02 .025 -.232 -2.960 .004 .956 1.046
-3,62E-04 .000 -.089 -1.140 .257 .965 1.036
.175 .877 .023 .200 .842 .453 2.208
(Constant) MSDN CAR BOPO LDR OWNER Model
1
B Std. Error Unstandardized
Coefficients
Beta Standardi
zed Coefficien
ts
t Sig. Tolerance VIF
Collinearity Statistics
Dependent Variable: ROA a.
Uji Asumsi Multikolinieritas
Lampiran 5
Uji Heteroskedastisitas
Warnings
For models with dependent variable LNE2, the following variables are constants or have missing correlations: LNOWNER. They will be deleted from the analysis.
Variables Entered/Removedb
LNLDR, LNMSDN, LNBOPO, LNCARa
. Enter
Removed Method
All requested variables entered. a.
Dependent Variable: LNE2 b.
Model Summary
.590a .349 .112 1.6581
Model 1
R R Square
Adjusted R Square
Std. Error of the Estimate
Predictors: (Constant), LNLDR, LNMSDN, LNBOPO, LNCAR
a.
ANOVAb
16.178 4 4.045 1.471 .276a
30.241 11 2.749
46.420 15
Regression
Squares df Mean Square F Sig.
Predictors: (Constant), LNLDR, LNMSDN, LNBOPO, LNCAR a.
Coefficientsa
-51.564 27.666 -1.864 .089
1.679 .823 .684 2.041 .066
4.752 2.283 .737 2.081 .062
6.253 4.991 .338 1.253 .236
.863 .983 .219 .878 .399
(Constant) LNMSDN LNCAR LNBOPO LNLDR Model
1
B Std. Error Unstandardized
Coefficients
Beta Standardi
zed Coefficien
ts
t Sig.
Dependent Variable: LNE2 a.
Uji Heterokedastisitas
Pengujian heterokedastisitas dilakukan dengan menggunakan metode Park, dengan membangun model Ln_e2 = b0 +
b1 X1 ... + b5 x5. Dari hasil analisis sebagaimana tampak pada tabel diatas, dapat diketahui bahwa variabel LNMSDN, LNCAR, LNBOPO, LN LDR tidak berpengaruh terhadap Ln_e2 (tingkat signifikansi > 0,05). Hal ini menandakan bahwa
model memenuhi asumsi heteroskedastisitas.