0% found this document useful (0 votes)
18 views6 pages

Panel EGLS Analysis Results

The document presents results from various econometric models (FEM, CEM, REM) analyzing a dependent variable Y over a balanced panel of 70 observations from 2014 to 2020. Key findings include significant coefficients for independent variables X1, X2, and X3 across different methods, with high R-squared values indicating strong model fit. Additionally, tests for heteroskedasticity and cross-sectional correlation suggest that the residuals exhibit heteroskedasticity and correlation among cross-sections.

Uploaded by

Aqilla Haya
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
18 views6 pages

Panel EGLS Analysis Results

The document presents results from various econometric models (FEM, CEM, REM) analyzing a dependent variable Y over a balanced panel of 70 observations from 2014 to 2020. Key findings include significant coefficients for independent variables X1, X2, and X3 across different methods, with high R-squared values indicating strong model fit. Additionally, tests for heteroskedasticity and cross-sectional correlation suggest that the residuals exhibit heteroskedasticity and correlation among cross-sections.

Uploaded by

Aqilla Haya
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

FEM

Dependent Variable: Y
Method: Panel EGLS (Cross-section weights)
Date: 07/23/22 Time: 10:00
Sample: 2014 2020
Periods included: 7
Cross-sections included: 10
Total panel (balanced) observations: 70
Linear estimation after one-step weighting matrix
Cross-section SUR (PCSE) standard errors & covariance (d.f. corrected)

Variable Coefficient Std. Error t-Statistic Prob.

C 7.154687 0.523250 13.67356 0.0000


X1 4.30E-06 7.61E-07 5.643545 0.0000
X2 0.436409 0.061008 7.153359 0.0000
X3 0.000102 1.90E-05 5.374794 0.0000

Effects Specification

Cross-section fixed (dummy variables)

Weighted Statistics

R-squared 0.984416 Mean dependent var 25.68465


Adjusted R-squared 0.981136 S.D. dependent var 18.49335
S.E. of regression 0.168025 Sum squared resid 1.609249
F-statistic 300.0586 Durbin-Watson stat 2.589463
Prob(F-statistic) 0.000000

Unweighted Statistics

R-squared 0.937010 Mean dependent var 12.70473


Sum squared resid 1.917422 Durbin-Watson stat 2.646494

CEM

Dependent Variable: Y
Method: Panel Least Squares
Date: 07/23/22 Time: 09:56
Sample: 2014 2020
Periods included: 7
Cross-sections included: 10
Total panel (balanced) observations: 70

Variable Coefficient Std. Error t-Statistic Prob.

C 6.826157 0.665336 10.25971 0.0000


X1 6.73E-06 6.45E-07 10.43416 0.0000
X2 0.405467 0.067125 6.040449 0.0000
X3 0.000212 3.16E-05 6.721221 0.0000

R-squared 0.867249 Mean dependent var 12.70473


Adjusted R-squared 0.861215 S.D. dependent var 0.664197
S.E. of regression 0.247439 Akaike info criterion 0.100141
Sum squared resid 4.040926 Schwarz criterion 0.228627
Log likelihood 0.495051 Hannan-Quinn criter. 0.151177
F-statistic 143.7239 Durbin-Watson stat 1.879679
Prob(F-statistic) 0.000000

REM

Dependent Variable: Y
Method: Panel EGLS (Cross-section random effects)
Date: 07/23/22 Time: 09:59
Sample: 2014 2020
Periods included: 7
Cross-sections included: 10
Total panel (balanced) observations: 70
Swamy and Arora estimator of component variances

Variable Coefficient Std. Error t-Statistic Prob.

C 6.672808 0.598172 11.15533 0.0000


X1 5.67E-06 5.75E-07 9.876356 0.0000
X2 0.447414 0.060478 7.397933 0.0000
X3 0.000161 2.59E-05 6.222026 0.0000

Effects Specification
S.D. Rho

Cross-section random 0.076726 0.1571


Idiosyncratic random 0.177706 0.8429

Weighted Statistics

R-squared 0.794532 Mean dependent var 8.368366


Adjusted R-squared 0.785193 S.D. dependent var 0.475683
S.E. of regression 0.220466 Sum squared resid 3.207953
F-statistic 85.07281 Durbin-Watson stat 1.947367
Prob(F-statistic) 0.000000

Unweighted Statistics

R-squared 0.857939 Mean dependent var 12.70473


Sum squared resid 4.324311 Durbin-Watson stat 1.444637
Uji LM

- Estimasi varians residual persamaan ke-i

0,0259 0,0042 -0,0060 -0,0082 0,0258 -0,0381 -0,0055 0,0162 -0,0194 0,0018

0,0042 0,0013 -0,0021 -0,0027 0,0063 -0,0087 -0,0015 0,0040 -0,0067 -35125567272,3816

-0,0060 -0,0021 0,0060 0,0066 -0,0078 0,0115 0,0024 -0,0055 0,0161 0,0021

-0,0082 -0,0027 0,0066 0,0078 -0,0107 0,0153 0,0032 -0,0077 0,0163 0,0020

0,0258 0,0063 -0,0078 -0,0107 0,0429 -0,0417 -0,0081 0,0261 -0,0384 0,0026

-0,0381 -0,0087 0,0115 0,0153 -0,0417 0,0801 0,0096 -0,0253 0,0340 -0,0028

-0,0055 -0,0015 0,0024 0,0032 -0,0081 0,0096 0,0019 -0,0052 0,0074 -1947038030,0602

0,0162 0,0040 -0,0055 -0,0077 0,0261 -0,0253 -0,0052 0,0165 -0,0222 0,0009

-0,0194 -0,0067 0,0161 0,0163 -0,0384 0,0340 0,0074 -0,0222 0,0730 0,0028

0,0018 -35125567272,3816 0,0021 0,0020 0,0026 -0,0028 -1947038030,0602 0,0009 0,0028 0,0018

- Estimasi varians residual model = 1,609249


- T = 7 (2014-2020)
- N = 10

[ ]
2
N i−1
σ^ 2i
- ∑∑ σ^
2
−1 = 10
i=1 j=1

[ ]
2
T
N i−1
σ^ 2i
- LM = ∑ ∑ 2 −1 = 10 ×3,5 = 35
2 i=1 j =1 σ^
- χ 2(N −1 ) = χ 2(9) = 16,91898
- Keputusan: Tolak H0
- Kesimpulan: Dengan tingkat signifikansi 5 persen dapat disimpulkan bahwa struktur
varians-kovarians residual bersifat heteroskedastis.
Uji λ LM

- Residual coefficient correlation antara persamaan ke-i dan ke-j

1,0000 0,5324 0,2298 0,3342 0,5968 0,6993 0,6230 0,6150 0,1995 0,0720
0,5324 1,0000 0,5643 0,7057 0,7233 0,7295 0,9282 0,7454 0,4785 0,0005
0,2298 0,5643 1,0000 0,9344 0,2390 0,2787 0,5049 0,3084 0,5923 0,3973
0,3342 0,7057 0,9344 1,0000 0,3450 0,3766 0,6953 0,4557 0,4653 0,2791
0,5968 0,7233 0,2390 0,3450 1,0000 0,5064 0,8132 0,9612 0,4712 0,0836
0,6993 0,7295 0,2787 0,3766 0,5064 1,0000 0,6168 0,4852 0,1976 0,0543
0,6230 0,9282 0,5049 0,6953 0,8132 0,6168 1,0000 0,8896 0,4047 0,0000
0,6150 0,7454 0,3084 0,4557 0,9612 0,4852 0,8896 1,0000 0,4088 0,0297
0,1995 0,4785 0,5923 0,4653 0,4712 0,1976 0,4047 0,4088 1,0000 0,0575
0,0720 0,0005 0,3973 0,2791 0,0836 0,0543 0,0000 0,0297 0,0575 1,0000

N i−1
T ∑ ∑ r ij = 144,4066
2
-
i=2 j=1
2
- χ N (N −1 ) = χ 2(45) = 61,65623338
( )
2

- Keputusan: Tolak H0
- Kesimpulan: Dengan tingkat signifikansi 5 persen dapat disimpulkan bahwa struktur
varians-kovarians residual bersifat heteroskedastis dan ada cross sectional correlation

REVISI

UJI CHOW

Redundant Fixed Effects Tests


Equation: Untitled
Test cross-section fixed effects

Effects Test Statistic d.f. Prob.

Cross-section F 18.942873 (9,58) 0.0000


Cross-section Chi-square 95.972191 9 0.0000

Cross-section fixed effects test equation:


Dependent Variable: LOGY
Method: Panel Least Squares
Date: 07/28/22 Time: 21:55
Sample: 2014 2020
Periods included: 7
Cross-sections included: 10
Total panel (balanced) observations: 70

Variable Coefficient Std. Error t-Statistic Prob.

C 2.056727 0.774823 2.654448 0.0099


LOGX1 0.433787 0.074555 5.818367 0.0000
LOGX2 0.528284 0.092066 5.738110 0.0000

R-squared 0.744397 Mean dependent var 12.70473


Adjusted R-squared 0.736767 S.D. dependent var 0.664197
S.E. of regression 0.340775 Akaike info criterion 0.726722
Sum squared resid 7.780543 Schwarz criterion 0.823086
Log likelihood -22.43528 Hannan-Quinn criter. 0.764999
F-statistic 97.56244 Durbin-Watson stat 0.772593
Prob(F-statistic) 0.000000

Uji Hausman

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 10.963743 2 0.0042

Cross-section random effects test comparisons:

Variable Fixed Random Var(Diff.) Prob.

LOGX1 0.140551 0.184723 0.000184 0.0011


LOGX2 0.552371 0.586739 0.001038 0.2862

Cross-section random effects test equation:


Dependent Variable: LOGY
Method: Panel Least Squares
Date: 07/28/22 Time: 21:57
Sample: 2014 2020
Periods included: 7
Cross-sections included: 10
Total panel (balanced) observations: 70

Variable Coefficient Std. Error t-Statistic Prob.

C 5.141137 0.928763 5.535466 0.0000


LOGX1 0.140551 0.052509 2.676690 0.0097
LOGX2 0.552371 0.081367 6.788605 0.0000
Effects Specification

Cross-section fixed (dummy variables)

R-squared 0.935116 Mean dependent var 12.70473


Adjusted R-squared 0.922811 S.D. dependent var 0.664197
S.E. of regression 0.184534 Akaike info criterion -0.387166
Sum squared resid 1.975052 Schwarz criterion -0.001710
Log likelihood 25.55082 Hannan-Quinn criter. -0.234058
F-statistic 75.99163 Durbin-Watson stat 2.456762
Prob(F-statistic) 0.000000

FEM SUR

Dependent Variable: LOGY


Method: Panel EGLS (Cross-section weights)
Date: 07/28/22 Time: 22:19
Sample: 2014 2020
Periods included: 7
Cross-sections included: 10
Total panel (balanced) observations: 70
Linear estimation after one-step weighting matrix
Cross-section SUR (PCSE) standard errors & covariance (d.f. corrected)

Variable Coefficient Std. Error t-Statistic Prob.

C 3.585699 0.234995 15.25863 0.0000


LOGX1 0.237291 0.049094 4.833376 0.0000
LOGX2 0.594269 0.059740 9.947621 0.0000

Effects Specification

Cross-section fixed (dummy variables)

Weighted Statistics

R-squared 0.983037 Mean dependent var 24.61940


Adjusted R-squared 0.979820 S.D. dependent var 16.07252
S.E. of regression 0.166908 Sum squared resid 1.615779
F-statistic 305.5595 Durbin-Watson stat 2.362668
Prob(F-statistic) 0.000000

Unweighted Statistics

R-squared 0.930162 Mean dependent var 12.70473


Sum squared resid 2.125851 Durbin-Watson stat 2.316346

You might also like