Negative Arch effects
Posted: Sat Oct 27, 2018 2:42 am
Hi Tom,
What is the reason behind getting a negative arch coefficient and how to solve it in CC-GARCH model?
MV-CC GARCH - Estimation by BFGS
Convergence in 46 Iterations. Final criterion was 0.0000004 <= 0.0000100
Daily(5) Data From 2007:11:15 To 2015:04:24
Usable Observations 1942
Log Likelihood -9740.0192
Variable Coeff Std Error T-Stat Signif
************************************************************************************
Mean Model(SP500RET)
1. Constant 0.029181968 0.018717737 1.55905 0.11898354
2. SP500RET{1} -0.050121919 0.020008320 -2.50505 0.01224327
3. GOLDRET{1} -0.010439866 0.017314504 -0.60295 0.54653867
4. WTIRET{1} 0.004746556 0.009825994 0.48306 0.62905228
Mean Model(GOLDRET)
5. Constant 0.027925514 0.026166610 1.06722 0.28587277
6. SP500RET{1} 0.007871211 0.017668418 0.44550 0.65596128
7. GOLDRET{1} -0.029456337 0.024216654 -1.21637 0.22384518
8. WTIRET{1} 0.055575514 0.010473750 5.30617 0.00000011
Mean Model(WTIRET)
9. Constant -0.003089988 0.034495644 -0.08958 0.92862401
10. SP500RET{1} 0.086881689 0.028684261 3.02890 0.00245448
11. GOLDRET{1} 0.027306429 0.031521966 0.86627 0.38634394
12. WTIRET{1} -0.058387791 0.019181103 -3.04403 0.00233434
13. C(1) 0.021491929 0.003248616 6.61572 0.00000000
14. C(2) 0.024232978 0.005598142 4.32875 0.00001500
15. C(3) 0.036341444 0.010700404 3.39627 0.00068311
16. A(1) -0.030315497 0.008782807 -3.45169 0.00055709
17. A(2) 0.047562791 0.009563894 4.97316 0.00000066
18. A(3) 0.029007545 0.008381600 3.46086 0.00053845
19. B(1) 0.908159142 0.011201729 81.07312 0.00000000
20. B(2) 0.932594849 0.009558300 97.56911 0.00000000
21. B(3) 0.927460429 0.008554144 108.42235 0.00000000
22. D(1) 0.201533544 0.021733510 9.27294 0.00000000
23. D(2) 0.009618938 0.011160736 0.86186 0.38876722
24. D(3) 0.074232714 0.012582721 5.89958 0.00000000
25. R(2,1) 0.055130069 0.019802019 2.78406 0.00536826
26. R(3,1) 0.341740584 0.015943505 21.43447 0.00000000
27. R(3,2) 0.229299421 0.019022643 12.05403 0.00000000
Multivariate Q(15)= 140.43687
Significance Level as Chi-Squared(135)= 0.35681
Test for Multivariate ARCH
Statistic Degrees Signif
375.94 360 0.27085
Information Criteria
AIC 10.111
SBC 10.189
Hannan-Quinn 10.139
(log) FPE 10.111
Many thanks for you great efforts
What is the reason behind getting a negative arch coefficient and how to solve it in CC-GARCH model?
MV-CC GARCH - Estimation by BFGS
Convergence in 46 Iterations. Final criterion was 0.0000004 <= 0.0000100
Daily(5) Data From 2007:11:15 To 2015:04:24
Usable Observations 1942
Log Likelihood -9740.0192
Variable Coeff Std Error T-Stat Signif
************************************************************************************
Mean Model(SP500RET)
1. Constant 0.029181968 0.018717737 1.55905 0.11898354
2. SP500RET{1} -0.050121919 0.020008320 -2.50505 0.01224327
3. GOLDRET{1} -0.010439866 0.017314504 -0.60295 0.54653867
4. WTIRET{1} 0.004746556 0.009825994 0.48306 0.62905228
Mean Model(GOLDRET)
5. Constant 0.027925514 0.026166610 1.06722 0.28587277
6. SP500RET{1} 0.007871211 0.017668418 0.44550 0.65596128
7. GOLDRET{1} -0.029456337 0.024216654 -1.21637 0.22384518
8. WTIRET{1} 0.055575514 0.010473750 5.30617 0.00000011
Mean Model(WTIRET)
9. Constant -0.003089988 0.034495644 -0.08958 0.92862401
10. SP500RET{1} 0.086881689 0.028684261 3.02890 0.00245448
11. GOLDRET{1} 0.027306429 0.031521966 0.86627 0.38634394
12. WTIRET{1} -0.058387791 0.019181103 -3.04403 0.00233434
13. C(1) 0.021491929 0.003248616 6.61572 0.00000000
14. C(2) 0.024232978 0.005598142 4.32875 0.00001500
15. C(3) 0.036341444 0.010700404 3.39627 0.00068311
16. A(1) -0.030315497 0.008782807 -3.45169 0.00055709
17. A(2) 0.047562791 0.009563894 4.97316 0.00000066
18. A(3) 0.029007545 0.008381600 3.46086 0.00053845
19. B(1) 0.908159142 0.011201729 81.07312 0.00000000
20. B(2) 0.932594849 0.009558300 97.56911 0.00000000
21. B(3) 0.927460429 0.008554144 108.42235 0.00000000
22. D(1) 0.201533544 0.021733510 9.27294 0.00000000
23. D(2) 0.009618938 0.011160736 0.86186 0.38876722
24. D(3) 0.074232714 0.012582721 5.89958 0.00000000
25. R(2,1) 0.055130069 0.019802019 2.78406 0.00536826
26. R(3,1) 0.341740584 0.015943505 21.43447 0.00000000
27. R(3,2) 0.229299421 0.019022643 12.05403 0.00000000
Multivariate Q(15)= 140.43687
Significance Level as Chi-Squared(135)= 0.35681
Test for Multivariate ARCH
Statistic Degrees Signif
375.94 360 0.27085
Information Criteria
AIC 10.111
SBC 10.189
Hannan-Quinn 10.139
(log) FPE 10.111
Many thanks for you great efforts