how to summarize VECH coefficients MVGARCH-BEKK Model

Discussions of ARCH, GARCH, and related models

how to summarize VECH coefficients MVGARCH-BEKK Model

Dear All,

I am trying to replicate Example 5.3 GARCH course material. I want to summarize VECH coeffcients for a bivariate model in the code.
Code: Select all
`summarize(title="(1,1) on (1,1)") %beta(10)ˆ2summarize(title="(1,2) on (1,1)") %beta(10)*%beta(13)*2.0summarize(title="(2,2) on (1,1)") %beta(13)ˆ2`

Would you please guide me how to rewrite the code?

Thanks and Regards,
Upananda
upani

Posts: 57
Joined: Wed Jun 25, 2014 3:31 am

Re: how to summarize VECH coefficients MVGARCH-BEKK Model

The 2nd edition has a cleaner implementation of that calculation. We sent you the update.
TomDoan

Posts: 7236
Joined: Wed Nov 01, 2006 5:36 pm

Re: how to summarize VECH coefficients MVGARCH-BEKK Model

Dear Sir,

Without Estima, my understanding of time series would have been incomplete.

With regards,
UPANANDA
upani

Posts: 57
Joined: Wed Jun 25, 2014 3:31 am

Re: how to summarize VECH coefficients MVGARCH-BEKK Model

Dear Sir,

I am trying run a multi variate garch model between mcx comex silver spot price. I got the following result. Would you please guide me on the following results?

Statistics on Series RMCX
Observations 16
Sample Mean 0.217507 Variance 5.936588
Standard Error 2.436511 SE of Sample Mean 0.609128
t-Statistic (Mean=0) 0.357079 Signif Level (Mean=0) 0.726007
Skewness -0.645381 Signif Level (Sk=0) 0.339815
Kurtosis (excess) 0.170232 Signif Level (Ku=0) 0.912341
Jarque-Bera 1.130029 Signif Level (JB=0) 0.568352

Statistics on Series RMCX
Observations 16
Sample Mean 0.217507 Variance 5.936588
Standard Error 2.436511 SE of Sample Mean 0.609128
t-Statistic (Mean=0) 0.357079 Signif Level (Mean=0) 0.726007
Skewness -0.645381 Signif Level (Sk=0) 0.339815
Kurtosis (excess) 0.170232 Signif Level (Ku=0) 0.912341
Jarque-Bera 1.130029 Signif Level (JB=0) 0.568352

MV-GARCH, BEKK - Estimation by BFGS
NO CONVERGENCE IN 58 ITERATIONS. FINAL NORMED GRADIENT 2.33786e+06
ESTIMATION POSSIBLY HAS STALLED OR MACHINE ROUNDOFF IS MAKING FURTHER PROGRESS DIFFICULT
TRY DIFFERENT SETTING FOR EXACTLINE, DERIVES OR ALPHA ON NLPAR
RESTARTING ESTIMATION FROM LAST ESTIMATES OR DIFFERENT INITIAL GUESSES/PMETHOD OPTION MIGHT ALSO WORK

With Heteroscedasticity/Misspecification Adjusted Standard Errors
Usable Observations 15
Log Likelihood -37.3176

Variable Coeff Std Error T-Stat Signif
************************************************************************************
Mean Model(RMCX)
1. RMCX{1} 0.858222426 0.070784712 12.12440 0.00000000
2. RCOMEX{1} -0.546144483 0.108587684 -5.02953 0.00000049
3. RSPOT{1} -0.221792786 0.043358342 -5.11534 0.00000031
4. Constant 0.501012906 0.140715982 3.56045 0.00037021
Mean Model(RCOMEX)
5. RMCX{1} 1.571676715 0.185533891 8.47110 0.00000000
6. RCOMEX{1} -0.903687810 0.109837104 -8.22753 0.00000000
7. RSPOT{1} -0.516110275 0.081595947 -6.32519 0.00000000
8. Constant 0.896708868 0.302989426 2.95954 0.00308100
Mean Model(RSPOT)
9. RMCX{1} 0.093766318 0.029989886 3.12660 0.00176842
10. RCOMEX{1} 0.388176986 0.068001932 5.70832 0.00000001
11. RSPOT{1} -0.540921536 0.027991856 -19.32425 0.00000000
12. Constant 0.773244827 0.225869639 3.42341 0.00061840

13. C(1,1) 1.016219634 0.433493288 2.34426 0.01906503
14. C(2,1) 2.273504603 0.628325075 3.61836 0.00029648
15. C(2,2) -0.005618038 0.758550040 -0.00741 0.99409069
16. C(3,1) 1.714770376 0.274004616 6.25818 0.00000000
17. C(3,2) -0.008775959 1.197373404 -0.00733 0.99415208
18. C(3,3) 0.000009540 0.004929692 0.00194 0.99845585
19. A(1,1) 0.189830482 0.044537655 4.26225 0.00002024
20. A(1,2) 0.729047376 0.225053599 3.23944 0.00119765
21. A(1,3) -1.042341373 0.154143384 -6.76215 0.00000000
22. A(2,1) 0.410824364 0.035538314 11.56004 0.00000000
23. A(2,2) 0.333231498 0.039568468 8.42164 0.00000000
24. A(2,3) 0.127290760 0.175225628 0.72644 0.46756958
25. A(3,1) 0.527180417 0.168923290 3.12083 0.00180344
26. A(3,2) 0.475946332 0.090445342 5.26225 0.00000014
27. A(3,3) 1.272522864 0.150569845 8.45138 0.00000000
28. B(1,1) -0.123866312 0.103609666 -1.19551 0.23188813
29. B(1,2) -0.127237004 0.020081686 -6.33597 0.00000000
30. B(1,3) 0.026886142 0.192570985 0.13962 0.88896278
31. B(2,1) 0.182596022 0.020155541 9.05935 0.00000000
32. B(2,2) 0.342087264 0.067492415 5.06853 0.00000040
33. B(2,3) 0.203623943 0.162363584 1.25412 0.20979722
34. B(3,1) -0.076985838 0.125707871 -0.61242 0.54026084
35. B(3,2) -0.027213726 0.153773516 -0.17697 0.85952978
36. B(3,3) 0.098151556 0.016472284 5.95859 0.00000000

Test of Exogeneity in Mean of All Variables
Chi-Squared(6)= 17257.778869 or F(6,*)= 2876.29648 with Significance Level 0.00000000

Test of Exogeneity in Mean of MCX Future
Chi-Squared(2)= 456.068940 or F(2,*)= 228.03447 with Significance Level 0.00000000

Test of Exogeneity in Mean of COMEX Future
Chi-Squared(2)= 88.924967 or F(2,*)= 44.46248 with Significance Level 0.00000000

Test of Exogeneity in Mean of LBMA Spot
Chi-Squared(2)= 46.429722 or F(2,*)= 23.21486 with Significance Level 0.00000000

Wald Test of Diagonal BEKK
Chi-Squared(12)=38044524.987879 or F(12,*)=3170377.08232 with Significance Level 0.00000000

Block Exclusion Test, MCX Future variance
Chi-Squared(4)= 579.785035 or F(4,*)= 144.94626 with Significance Level 0.00000000

Block Exclusion Test, COMEX Future Variance
Chi-Squared(4)= 47216.009134 or F(4,*)= 11804.00228 with Significance Level 0.00000000

Block Exclusion Test, LBMA spot Variance
Chi-Squared(4)= 8522.905524 or F(4,*)= 2130.72638 with Significance Level 0.00000000

MV-GARCH, BEKK - Estimation by BFGS
NO CONVERGENCE IN 58 ITERATIONS. FINAL NORMED GRADIENT 2.33786e+06
ESTIMATION POSSIBLY HAS STALLED OR MACHINE ROUNDOFF IS MAKING FURTHER PROGRESS DIFFICULT
TRY DIFFERENT SETTING FOR EXACTLINE, DERIVES OR ALPHA ON NLPAR
RESTARTING ESTIMATION FROM LAST ESTIMATES OR DIFFERENT INITIAL GUESSES/PMETHOD OPTION MIGHT ALSO WORK

With Heteroscedasticity/Misspecification Adjusted Standard Errors
Usable Observations 15
Log Likelihood -37.3176

Variable Coeff Std Error T-Stat Signif
************************************************************************************
Mean Model(RMCX)
1. RMCX{1} 0.858222426 0.070784712 12.12440 0.00000000
2. RCOMEX{1} -0.546144483 0.108587684 -5.02953 0.00000049
3. RSPOT{1} -0.221792786 0.043358342 -5.11534 0.00000031
4. Constant 0.501012906 0.140715982 3.56045 0.00037021
Mean Model(RCOMEX)
5. RMCX{1} 1.571676715 0.185533891 8.47110 0.00000000
6. RCOMEX{1} -0.903687810 0.109837104 -8.22753 0.00000000
7. RSPOT{1} -0.516110275 0.081595947 -6.32519 0.00000000
8. Constant 0.896708868 0.302989426 2.95954 0.00308100
Mean Model(RSPOT)
9. RMCX{1} 0.093766318 0.029989886 3.12660 0.00176842
10. RCOMEX{1} 0.388176986 0.068001932 5.70832 0.00000001
11. RSPOT{1} -0.540921536 0.027991856 -19.32425 0.00000000
12. Constant 0.773244827 0.225869639 3.42341 0.00061840

13. C(1,1) 1.016219634 0.433493288 2.34426 0.01906503
14. C(2,1) 2.273504603 0.628325075 3.61836 0.00029648
15. C(2,2) -0.005618038 0.758550040 -0.00741 0.99409069
16. C(3,1) 1.714770376 0.274004616 6.25818 0.00000000
17. C(3,2) -0.008775959 1.197373404 -0.00733 0.99415208
18. C(3,3) 0.000009540 0.004929692 0.00194 0.99845585
19. A(1,1) 0.189830482 0.044537655 4.26225 0.00002024
20. A(1,2) 0.729047376 0.225053599 3.23944 0.00119765
21. A(1,3) -1.042341373 0.154143384 -6.76215 0.00000000
22. A(2,1) 0.410824364 0.035538314 11.56004 0.00000000
23. A(2,2) 0.333231498 0.039568468 8.42164 0.00000000
24. A(2,3) 0.127290760 0.175225628 0.72644 0.46756958
25. A(3,1) 0.527180417 0.168923290 3.12083 0.00180344
26. A(3,2) 0.475946332 0.090445342 5.26225 0.00000014
27. A(3,3) 1.272522864 0.150569845 8.45138 0.00000000
28. B(1,1) -0.123866312 0.103609666 -1.19551 0.23188813
29. B(1,2) -0.127237004 0.020081686 -6.33597 0.00000000
30. B(1,3) 0.026886142 0.192570985 0.13962 0.88896278
31. B(2,1) 0.182596022 0.020155541 9.05935 0.00000000
32. B(2,2) 0.342087264 0.067492415 5.06853 0.00000040
33. B(2,3) 0.203623943 0.162363584 1.25412 0.20979722
34. B(3,1) -0.076985838 0.125707871 -0.61242 0.54026084
35. B(3,2) -0.027213726 0.153773516 -0.17697 0.85952978
36. B(3,3) 0.098151556 0.016472284 5.95859 0.00000000

0.036 0.156 0.169 0.200 0.433 0.278
0.138 0.363 0.137 0.475 0.371 0.251
0.532 0.486 0.111 0.694 0.317 0.227
-0.198 -0.404 0.052 -0.308 0.590 0.671
-0.760 -0.255 0.042 0.432 0.485 0.606
1.086 -0.265 0.016 -2.653 0.324 1.619

( 1.403,-0.000) ( 0.655,-1.110) ( 0.655, 1.110) (-0.181, 0.117) (-0.181,-0.117) ( 0.045, 0.000)

Multivariate Q Test
Test Run Over 2007 to 2021
Lags Tested 10
Degrees of Freedom 90
Q Statistic 90.10308
Signif Level 0.47712

## MAT14. Non-invertible Matrix. Using Generalized Inverse for SYMMETRIC.
The Error Occurred At Location 651, Line 59 of MVARCHTEST
C:\Users\Public\Documents\Estima\WinRATS Pro 10.0\mvarchtest.src Line 87

Summary of Function of Coefficients

Value 0.01534286 t-Statistic 0.59775
Standard Error 0.02566749 Signif Level 0.5500037

Summary of Function of Coefficients

Value -0.0452350 t-Statistic -1.06924
Standard Error 0.0423059 Signif Level 0.2849637

Summary of Function of Coefficients

Value 0.03334131 t-Statistic 4.52967
Standard Error 0.00736064 Signif Level 0.0000059

Summary of Function of Coefficients

Value 0.01534286 t-Statistic 0.59775
Standard Error 0.02566749 Signif Level 0.5500037

Summary of Function of Coefficients

Value -0.0452350 t-Statistic -1.06924
Standard Error 0.0423059 Signif Level 0.2849637

Summary of Function of Coefficients

Value 0.03334131 t-Statistic 4.52967
Standard Error 0.00736064 Signif Level 0.0000059

1,1 2,1 2,2 3,1 3,2 3,3
1,1 0.0153 -0.0452 0.0333 0.0191 -0.0281 0.0059
( 0.0257) ( 0.0423) ( 0.0074) ( 0.0471) ( 0.0487) ( 0.0194)
2,1 0.0158 -0.0656 0.0625 0.0132 -0.0313 0.0021
( 0.0157) ( 0.0332) ( 0.0076) ( 0.0394) ( 0.0664) ( 0.0153)
2,2 0.0162 -0.0871 0.1170 0.0069 -0.0186 0.0007
( 0.0051) ( 0.0041) ( 0.0462) ( 0.0402) ( 0.1016) ( 0.0084)
3,1 -0.0033 -0.0203 0.0372 -0.0142 0.0022 -0.0076
( 0.0266) ( 0.0347) ( 0.0260) ( 0.0304) ( 0.0089) ( 0.0136)
3,2 -0.0034 -0.0167 0.0697 -0.0132 0.0280 -0.0027
( 0.0250) ( 0.0806) ( 0.0692) ( 0.0134) ( 0.0280) ( 0.0155)
3,3 0.0007 0.0109 0.0415 0.0053 0.0400 0.0096
( 0.0104) ( 0.0697) ( 0.0661) ( 0.0387) ( 0.0252) ( 0.0032)

I am having doubt on convergence issues and is there any method i can try over here.

Looking forward to your reply on the same

With sincere regards,
Upananda Pani
upani

Posts: 57
Joined: Wed Jun 25, 2014 3:31 am

Re: how to summarize VECH coefficients MVGARCH-BEKK Model

You are showing only 16 observations. There's something wrong with your data or how you are getting your data in.
TomDoan

Posts: 7236
Joined: Wed Nov 01, 2006 5:36 pm

Re: how to summarize VECH coefficients MVGARCH-BEKK Model

Dear Sir,

Thank you very much for your reply. I have attached the data set for your reference.

With sincere regards,
Upananda Pani
Attachments
silvermcx.xlsx
(43.48 KiB) Downloaded 27 times
upani

Posts: 57
Joined: Wed Jun 25, 2014 3:31 am

Return to ARCH and GARCH Models

Who is online

Users browsing this forum: No registered users and 2 guests