BAYU HERMAWAN

143210136
Estimate Model CEM
Dependent Variable: TK
Method: Panel Least Squares
Date: 09/29/23 Time: 07:17
Sample: 2010 2017
Periods included: 8
Cross-sections included: 10
Total panel (balanced) observations: 80

Variable Coefficient Std. Error t-Statistic Prob.

C 6.009394 0.609277 9.863154 0.0000

CR 1.132786 0.114824 9.865420 0.0000
DAR -0.077481 0.009178 -8.441959 0.0000

Root MSE 1.244129 R-squared 0.840600

Mean dependent var 4.740473 Adjusted R-squared 0.836460
S.D. dependent var 3.135834 S.E. of regression 1.268133
Akaike info criterion 3.349748 Sum squared resid 123.8285
Schwarz criterion 3.439074 Log likelihood -130.9899
Hannan-Quinn criter. 3.385561 F-statistic 203.0312
Durbin-Watson stat 0.364325 Prob(F-statistic) 0.000000

Estimate Model FEM

Dependent Variable: TK
Method: Panel Least Squares
Date: 09/29/23 Time: 07:18
Sample: 2010 2017
Periods included: 8
Cross-sections included: 10
Total panel (balanced) observations: 80

Variable Coefficient Std. Error t-Statistic Prob.

C 7.205720 0.885596 8.136580 0.0000

CR 0.908848 0.165805 5.481435 0.0000
DAR -0.093491 0.014675 -6.370805 0.0000

Effects Specification

Cross-section fixed (dummy variables)

Root MSE 0.844294 R-squared 0.926592

Mean dependent var 4.740473 Adjusted R-squared 0.914717
S.D. dependent var 3.135834 S.E. of regression 0.915766
Akaike info criterion 2.799368 Sum squared resid 57.02661
Schwarz criterion 3.156672 Log likelihood -99.97474
Hannan-Quinn criter. 2.942622 F-statistic 78.02975
Durbin-Watson stat 0.764662 Prob(F-statistic) 0.000000

Estimate Model REM

Dependent Variable: TK
Method: Panel EGLS (Cross-section random effects)
Date: 09/29/23 Time: 07:19
Sample: 2010 2017
Periods included: 8
Cross-sections included: 10
Total panel (balanced) observations: 80
Swamy and Arora estimator of component variances

Variable Coefficient Std. Error t-Statistic Prob.

C 6.863699 0.841760 8.153991 0.0000

CR 0.970826 0.146719 6.616919 0.0000
DAR -0.088823 0.012525 -7.091638 0.0000

Effects Specification
S.D. Rho

Cross-section random 1.022271 0.5548

Idiosyncratic random 0.915766 0.4452

Weighted Statistics

Root MSE 0.891451 R-squared 0.708050

Mean dependent var 1.431322 Adjusted R-squared 0.700467
S.D. dependent var 1.660255 S.E. of regression 0.908651
Sum squared resid 63.57474 F-statistic 93.37206
Durbin-Watson stat 0.678891 Prob(F-statistic) 0.000000

Unweighted Statistics

R-squared 0.835955 Mean dependent var 4.740473

Sum squared resid 127.4368 Durbin-Watson stat 0.338680

Memilih model dengan menggunakan Uji Chow, Uji Hausman, dan Uji Lagrange
Multiplier(LM).
1. Uji Chow : Pooled Least Square (Common Effect Model) vs Fixed Effect Model (FEM)
Redundant Fixed Effects Tests
Equation: Untitled
Test cross-section fixed e ffects

Effects Test Statistic d.f. Prob.

Cross-section F 8.850696 (9,68) 0.0000

Cross-section Chi-square 62.030344 9 0.0000
Cross-section fixed effects test equation:
Dependent Variable: TK
Method: Panel Least Squares
Date: 09/29/23 Time: 07:24
Sample: 2010 2017
Periods included: 8
Cross-sections included: 10
Total panel (balanced) observations: 80

Variable Coefficient Std. Error t-Statistic Prob.

C 6.009394 0.609277 9.863154 0.0000

CR 1.132786 0.114824 9.865420 0.0000
DAR -0.077481 0.009178 -8.441959 0.0000

Root MSE 1.244129 R-squared 0.840600

Mean dependent var 4.740473 Adjusted R-squared 0.836460
S.D. dependent var 3.135834 S.E. of regression 1.268133
Akaike info criterion 3.349748 Sum squared resid 123.8285
Schwarz criterion 3.439074 Log likelihood -130.9899
Hannan-Quinn criter. 3.385561 F-statistic 203.0312
Durbin-Watson stat 0.364325 Prob(F-statistic) 0.000000

Hasil redundant fixed effects atau likelihood ratio untuk model memiliki nilai probabilitas
F lebih kecil dari alpha (0,05), sehingga H0 ditolak dan H1 diterima, model yang sesuai dari hasil
ini yaitu fixed effects (Karena nilai probabilitas F sebesar 0.0000 < 0,05). Karena yang terpilih
adalah FEM maka uji dilanjutkan ke Uji Hausman.

2. Uji Hausman : Fixed Effect Model vs Random Effect Model

Correlated Random Effects - Hausman Test

Equation: Untitled
Test cross-section random effects

Chi-Sq.
Test Summary Statistic Chi-Sq. d.f. Prob.

Cross-section random 0.808155 2 0.6676

Cross-section random effects test comparisons:

Variable Fixed Random Var(Diff.) Prob.

CR 0.908848 0.970826 0.005965 0.4223

DAR -0.093491 -0.088823 0.000058 0.5416

Cross-section random effects test equation:

Dependent Variable: TK
Method: Panel Least Squares
Date: 09/29/23 Time: 07:27
Sample: 2010 2017
Periods included: 8
Cross-sections included: 10
Total panel (balanced) observations: 80

Variable Coefficient Std. Error t-Statistic Prob.

C 7.205720 0.885596 8.136580 0.0000

CR 0.908848 0.165805 5.481435 0.0000
DAR -0.093491 0.014675 -6.370805 0.0000

Effects Specification

Cross-section fixed (dummy variables)

Root MSE 0.844294 R-squared 0.926592

Mean dependent var 4.740473 Adjusted R-squared 0.914717
S.D. dependent var 3.135834 S.E. of regression 0.915766
Akaike info criterion 2.799368 Sum squared resid 57.02661
Schwarz criterion 3.156672 Log likelihood -99.97474
Hannan-Quinn criter. 2.942622 F-statistic 78.02975
Durbin-Watson stat 0.764662 Prob(F-statistic) 0.000000

Hasil redundant fixed effects atau likelihood ratio untuk model memiliki nilai probabilitas
F lebih besar dari alpha (0,05), sehingga H1 ditolak dan H0 diterima, model yang sesuai dari hasil
ini yaitu REM (Karena nilai probabilitas F sebesar 0.6676 > 0,0000). Karena REM yang terpilih
maka uji dilanjutkan ke UJI LM
3. Uji Lagrange Multiplier (LM): Random Effect Model vs Common Effect Model
Lagrange Multiplier Tests for Random Effects
Null hypotheses: No effects
Alternative hypotheses: Two-sided (Breusch-Pagan) and one-sided
(all others) alternatives

Test Hypothesis
Cross-section Time Both

Breusch-Pagan 58.51658 0.078496 58.59508

(0.0000) (0.7793) (0.0000)

Honda 7.649613 0.280172 5.607205

(0.0000) (0.3897) (0.0000)

King-Wu 7.649613 0.280172 5.269873

(0.0000) (0.3897) (0.0000)

Standardized Honda 9.361454 0.500211 3.392832

(0.0000) (0.3085) (0.0003)

Standardized King-Wu 9.361454 0.500211 2.987857

(0.0000) (0.3085) (0.0014)
Gourieroux, et al. -- -- 58.59508
(0.0000)

Jika cross section random kurang dari 5% atau lebih kecil dari 0,05 artinya menolak H0
dan H1 diterima model yang sesuai dari hasil ini yaitu REM.

Ringkasan Hasil Uji Chow, Hausman, Lagrange Multiplier:

PENGUJIAN HASIL KEPUTUSAN
UJI CHOW PROB. > 0,05 CEM
PROB. < 0,05 FEM
UJI HAUSMAN PROB. > 0,05 REM
PROB. < 0,05 FEM
UJI LM PROB. > 0,05 CEM
PROB. < 0,05 REM

Interpretasi Model Terpilih Penelitian Hasil Olah Data Panel EVIEWS (UJI ASUMSI
KLASIK)
Dependent Variable: TK
Method: Panel EGLS (Cross-section random effects)
Date: 09/29/23 Time: 07:19
Sample: 2010 2017
Periods included: 8
Cross-sections included: 10
Total panel (balanced) observations: 80
Swamy and Arora estimator of component variances

Variable Coefficient Std. Error t-Statistic Prob.

C 6.863699 0.841760 8.153991 0.0000

CR 0.970826 0.146719 6.616919 0.0000
DAR -0.088823 0.012525 -7.091638 0.0000

Effects Specification
S.D. Rho

Cross-section random 1.022271 0.5548

Idiosyncratic random 0.915766 0.4452

Weighted Statistics
Root MSE 0.891451 R-squared 0.708050
Mean dependent var 1.431322 Adjusted R-squared 0.700467
S.D. dependent var 1.660255 S.E. of regression 0.908651
Sum squared resid 63.57474 F-statistic 93.37206
Durbin-Watson stat 0.678891 Prob(F-statistic) 0.000000

Unweighted Statistics

R-squared 0.835955 Mean dependent var 4.740473

Sum squared resid 127.4368 Durbin-Watson stat 0.338680

Hasil uji t pada Tabel di atas menunjukkan tingkat signifikansi pengaruh variable CR dan
DAR terhadap variabel TK.

➢ Variabel CR memiliki nilai probabilitas 0,0000 < 0,05 sehingga variabel X1 memiliki pengaruh
signifikan terhadap variabel TK.

➢ Variabel DAR memiliki nilai probabilitas 0,0000 < 0,05 sehingga variabel X2 memiliki
pengaruh signifikan terhadap variabel TK.
Hasil uji F pada Tabel di atas menunjukkan nilai Prob (Fstatistic) 0,000000 < 0,05 sehingga
variabel X1 dan X2 secara bersama-sama memiliki pengaruh signifikan terhadap variabel Y.
Hasil pengolahan data menunjukkan bahwa koefisien determinasi (Adj. R2) sebesar 0,7005
(70,05%). Hal ini menggambarkan bahwa variabel-variabel independent(CR & DAR) secara
bersama-sama mampu memberi penjelasan mengenai variabel dependen sebesar 70,05 persen.
Adapun 29,95% lagi dijelaskan oleh variabel lain yang tidak dimasukkan dalam model penelitian
ini.