NAMA : DEFFRY APRIALDY

NIM : 190522030
MATA KULIAH : STATISTIKA PENELITIAN
SIFAT UJIAN : TAKE HOME
APLIKASI : EVIEW VERSI 10

1. Data
NO Y 2014 Y 2015 X1 2014 X2 2015 X3 2014 X4 2015
1 0,0716 0,0748 0,106 0,119 0,037 0,023
2 0,0435 0,0351 0,4562 0,0012 0,0694 0,0304
3 0,0148 0,136 0,0161 0,1011 0,1384 0,0493
4 0,0601 0,015 0,019 0,7613 0,6551 0,125
5 0,8629 0,0193 0,1293 0,8947 0,6254 0,1245
6 0,0739 0,1149 0,1919 0,1695 0,0725 0,0623
7 0,0127 0,045 0,0066 0,0188 0,0069 0,0181
8 0,143 0,0762 0,1018 0,0132 0,0929 0,1320
9 0,1008 0,2654 0,0275 0,037 0,0946 0,0022
10 0,042 0,0057 0,0233 0,1007 0,1715 0,0347
11 0,0135 0,007 0,0055 0,0068 0,0127 0,0069
12 0,0487 0,0693 0,0176 0,0019 0,0380 0,0071
13 0,0312 0,0749 0,0605 0,0486 0,0345 0,0519
14 0,1187 0,1024 0,3426 0,2138 0,6433 0,1784

2. Uji Normalitas Data

dapat dilihat bahwa nilai Probability sebesar 0,290640 > 0,05 artinya residual data penelitian
terdistribusi secara normal.

3. Uji Asumsi Klasik

 Common Effect Model
Dependent Variable: Y
Method: Panel Least Squares
Date: 01/20/20 Time: 02:36
Sample: 2014 2015
Periods included: 2
Cross-sections included: 14
Total panel (balanced) observations: 28

Variable Coefficient Std. Error t-Statistic Prob.  

C 0.052063 0.035711 1.457917 0.1573

X1 -0.093717 0.123608 -0.758175 0.4554
X2 0.450737 0.145545 3.096880 0.0048

R-squared 0.279767     Mean dependent var 0.095657

Adjusted R-squared 0.222149     S.D. dependent var 0.160234
S.E. of regression 0.141320     Akaike info criterion -0.974627
Sum squared resid 0.499281     Schwarz criterion -0.831891
Log likelihood 16.64478     Hannan-Quinn criter. -0.930991
F-statistic 4.855499     Durbin-Watson stat 1.824161
Prob(F-statistic) 0.016535

 Fixed Effect Model

Dependent Variable: Y
Method: Panel Least Squares
Date: 01/20/20 Time: 02:38
Sample: 2014 2015
Periods included: 2
Cross-sections included: 14
Total panel (balanced) observations: 28

Variable Coefficient Std. Error t-Statistic Prob.  

C 0.121384 0.068658 1.767953 0.1025

X1 -0.367146 0.223677 -1.641412 0.1266
X2 0.210601 0.296308 0.710751 0.4908

Effects Specification

Cross-section fixed (dummy variables)

R-squared 0.687106     Mean dependent var 0.095657

Adjusted R-squared 0.295989     S.D. dependent var 0.160234
S.E. of regression 0.134445     Akaike info criterion -0.879767
Sum squared resid 0.216905     Schwarz criterion -0.118507
Log likelihood 28.31674     Hannan-Quinn criter. -0.647042
F-statistic 1.756780     Durbin-Watson stat 3.733333
Prob(F-statistic) 0.165308

Kemudian kita melakukan Uji Chow untuk mengetahui model yang akan kita gunakan antara Common
Effect Model atau Fixed Effect Model terlihat hasil output dibawah ini :

Redundant Fixed Effects Tests

Equation: Untitled
Test cross-section fixed effects

Effects Test Statistic   d.f.  Prob. 

Cross-section F 1.201704 (13,12) 0.3783

Cross-section Chi-square 23.343920 13 0.0377

Cross-section fixed effects test equation:

Dependent Variable: Y
Method: Panel Least Squares
Date: 01/20/20 Time: 02:38
Sample: 2014 2015
Periods included: 2
Cross-sections included: 14
Total panel (balanced) observations: 28

Variable Coefficient Std. Error t-Statistic Prob.  

C 0.052063 0.035711 1.457917 0.1573

X1 -0.093717 0.123608 -0.758175 0.4554
X2 0.450737 0.145545 3.096880 0.0048

R-squared 0.279767     Mean dependent var 0.095657

Adjusted R-squared 0.222149     S.D. dependent var 0.160234
S.E. of regression 0.141320     Akaike info criterion -0.974627
Sum squared resid 0.499281     Schwarz criterion -0.831891
Log likelihood 16.64478     Hannan-Quinn criter. -0.930991
F-statistic 4.855499     Durbin-Watson stat 1.824161
Prob(F-statistic) 0.016535

Nilai Probabilitas Cross-section Chi-Square sebesar 0,0377 artinya kurang dari 0,05 maka Fixed Effect
Model adalah mode yang sesuai, oleh karena itu kita akan bandingkan dengan Random Effect Model
melalui Uji Hausman. Hasil output dapat dilihat dibawah ini

Dependent Variable: Y
Method: Panel EGLS (Cross-section random effects)
Date: 01/20/20 Time: 02:39
Sample: 2014 2015
Periods included: 2
Cross-sections included: 14
Total panel (balanced) observations: 28
Swamy and Arora estimator of component variances

Variable Coefficient Std. Error t-Statistic Prob.  

C 0.052063 0.033973 1.532468 0.1380

X1 -0.093717 0.117595 -0.796945 0.4330
X2 0.450737 0.138465 3.255241 0.0032

Effects Specification
S.D.   Rho  

Cross-section random 0.000000 0.0000

Idiosyncratic random 0.134445 1.0000

Weighted Statistics

R-squared 0.279767     Mean dependent var 0.095657

Adjusted R-squared 0.222149     S.D. dependent var 0.160234
S.E. of regression 0.141320     Sum squared resid 0.499281
F-statistic 4.855499     Durbin-Watson stat 1.824161
Prob(F-statistic) 0.016535

Unweighted Statistics

R-squared 0.279767     Mean dependent var 0.095657

Sum squared resid 0.499281     Durbin-Watson stat 1.824161

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 5.024196 2 0.0811

** WARNING: estimated cross-section random effects variance is zero.

Cross-section random effects test comparisons:

Variable Fixed   Random  Var(Diff.)  Prob. 

 X1 -0.367146 -0.093717 0.036203 0.1507
X2 0.210601 0.450737 0.068626 0.3593

Cross-section random effects test equation:

Dependent Variable: Y
Method: Panel Least Squares
Date: 01/20/20 Time: 02:40
Sample: 2014 2015
Periods included: 2
Cross-sections included: 14
Total panel (balanced) observations: 28

Variable Coefficient Std. Error t-Statistic Prob.  

C 0.121384 0.068658 1.767953 0.1025

X1 -0.367146 0.223677 -1.641412 0.1266
X2 0.210601 0.296308 0.710751 0.4908

Effects Specification

Cross-section fixed (dummy variables)

R-squared 0.687106     Mean dependent var 0.095657

Adjusted R-squared 0.295989     S.D. dependent var 0.160234
S.E. of regression 0.134445     Akaike info criterion -0.879767
Sum squared resid 0.216905     Schwarz criterion -0.118507
Log likelihood 28.31674     Hannan-Quinn criter. -0.647042
F-statistic 1.756780     Durbin-Watson stat 3.733333
Prob(F-statistic) 0.165308

Setelah dilakukan Uji Hausman terlihat Cross-section Random nilai probabilitasnya 0,0811 artinya lebih
dari 0,05 maka Uji Fixed Effect Model tidak sesuai.

Uji Multikolinearitas

X1 X2

X1  1.000000  0.133814
X2  0.133814  1.000000

dapat dilihat bahwa nillai korelasi antar variabel independen kurang dari 0,8 maka dapat disimpulkan
tidak terjadi Multikolinearitas.

Uji Heteroskedastisitas
Dependent Variable: RESABS
Method: Panel Least Squares
Date: 01/20/20 Time: 03:39
Sample: 2014 2015
Periods included: 2
Cross-sections included: 14
Total panel (balanced) observations: 28

Variable Coefficient Std. Error t-Statistic Prob.  

C 0.018561 0.013022 1.425409 0.1664

X1 0.145277 0.045073 3.223117 0.0035
X2 0.149075 0.053073 2.808876 0.0095

R-squared 0.457449     Mean dependent var 0.058108

Adjusted R-squared 0.414045     S.D. dependent var 0.067320
S.E. of regression 0.051532     Akaike info criterion -2.992280
Sum squared resid 0.066388     Schwarz criterion -2.849543
Log likelihood 44.89191     Hannan-Quinn criter. -2.948644
F-statistic 10.53931     Durbin-Watson stat 0.434527
Prob(F-statistic) 0.000479
 dapat dilihat bahwa nillai probabilitas masing-masing variabel independen kurang dari 0,05 maka dapat
disimpulkan terjadinya heteroskedastisitas.

Uji Asumsi Non Autokorelasi

Dependent Variable: Y
Method: Panel Least Squares
Date: 01/20/20 Time: 02:38
Sample: 2014 2015
Periods included: 2
Cross-sections included: 14
Total panel (balanced) observations: 28

Variable Coefficient Std. Error t-Statistic Prob.  

C 0.121384 0.068658 1.767953 0.1025

X1 -0.367146 0.223677 -1.641412 0.1266
X2 0.210601 0.296308 0.710751 0.4908

Effects Specification

Cross-section fixed (dummy variables)

R-squared 0.687106     Mean dependent var 0.095657

Adjusted R-squared 0.295989     S.D. dependent var 0.160234
S.E. of regression 0.134445     Akaike info criterion -0.879767
Sum squared resid 0.216905     Schwarz criterion -0.118507
Log likelihood 28.31674     Hannan-Quinn criter. -0.647042
F-statistic 1.756780     Durbin-Watson stat 3.733333
Prob(F-statistic) 0.165308

Dapat dilihat bahwa nilai Durbin Watson stat akan dibandingkan dengan dLl dan dU pada tabel
durbin watson dengan N 14 dan K 2, maka nilai dL 0,9054 dan dU 1,5507 karena nilai DW 3,7333
lebih besar dari dL 1,5507 maka dapat disimpulkan terjadinya autokorelasi negatif sehingga
asumsi non-autokorelasi terpenuhi.

4. Regresi Linear Berganda

Dependent Variable: Y
Method: Panel Least Squares
Date: 01/20/20 Time: 04:05
Sample: 2014 2015
Periods included: 2
Cross-sections included: 14
Total panel (balanced) observations: 28

Variable Coefficient Std. Error t-Statistic Prob.  

C 0.052063 0.035711 1.457917 0.1573

X1 -0.093717 0.123608 -0.758175 0.4554
X2 0.450737 0.145545 3.096880 0.0048

R-squared 0.279767     Mean dependent var 0.095657

Adjusted R-squared 0.222149     S.D. dependent var 0.160234
S.E. of regression 0.141320     Akaike info criterion -0.974627
Sum squared resid 0.499281     Schwarz criterion -0.831891
Log likelihood 16.64478     Hannan-Quinn criter. -0.930991
F-statistic 4.855499     Durbin-Watson stat 1.824161
Prob(F-statistic) 0.016535

Dapat dilihat bahwa :

  X1 nilai Probability sebesar 0,4554 < 0,05 artinya X1 berpengaruh terhadap Y
 X2 nilai Probability sebesar 0,0048 < 0,05 artinya X2 berpengaruh terhadap Y

Bila hasil penelitian menggunakan model Matematika maka, persamaannya adalah sebagai berikut :

Y = 0,0520 – 0,0937 + 0,4507