The RATS Software Forum

Please I need help with interpretation of results of fittingba DCC Garch (1,2) as seen below
MV-DCC GARCH - Estimation by BFGS
Convergence in 74 Iterations. Final criterion was 0.0000012 <= 0.0000100

Daily(7) Data From 2017:01:19 To 2018:01:17
Usable Observations 364
Log Likelihood -4308.1937

Variable Coeff Std Error T-Stat Signif
************************************************************************************
1. Mean(BTCR) 2.312074747 0.450373759 5.13368 0.00000028
2. Mean(ETHR) 1.742694354 0.595917425 2.92439 0.00345133
3. Mean(LTCR) -0.244303716 0.408283502 -0.59837 0.54959452

4. C(1) 2.272894830 2.266069944 1.00301 0.31585518
5. C(2) 4.706665120 3.968248083 1.18608 0.23559017
6. C(3) 0.546324349 0.774388192 0.70549 0.48050437
7. A{1}(1) 1.024716036 0.184444272 5.55569 0.00000003
8. A{1}(2) 1.186749728 0.135779917 8.74024 0.00000000
9. A{1}(3) 1.061465150 0.067339652 15.76285 0.00000000
10. A{2}(1) -0.807144411 0.154888604 -5.21113 0.00000019
11. A{2}(2) -0.933860038 0.073516858 -12.70267 0.00000000
12. A{2}(3) -0.939444122 0.051290315 -18.31621 0.00000000
13. B(1) 0.900754929 0.035582977 25.31421 0.00000000
14. B(2) 0.880589161 0.039909675 22.06455 0.00000000
15. B(3) 0.936317420 0.022807471 41.05310 0.00000000
16. DCC(A) 0.141483514 0.027079692 5.22471 0.00000017
17. DCC(B) 0.842412309 0.032755606 25.71811 0.00000000
18. Shape 2.890357456 0.244562491 11.81848 0.00000000


Covariance\Correlation Matrix of Coefficients
Mean(BTCR) Mean(ETHR) Mean(LTCR) C(1) C(2) C(3) A{1}(1) A{1}(2) A{1}(3) A{2}(1) A{2}(2) A{2}(3) B(1) B(2) B(3) DCC(A) DCC(B) Shape
Mean(BTCR) 0.20283652 0.26538635 0.43722296 0.00306088 0.00521557 -0.00762666 0.03024676 -0.01591394 -0.01783560 -0.03387423 0.01993340 0.02151804 0.00237705 0.01400812 0.00992806 -0.07262862 0.05041021 -0.03625577
Mean(ETHR) 0.07122587 0.35511758 0.39556998 0.00454472 0.14848637 0.05833824 0.10694335 0.06599121 0.06613297 -0.04839914 0.08535319 0.01073770 -0.07685613 -0.12062367 -0.07411338 0.18718877 -0.19202667 -0.11661117
Mean(LTCR) 0.08039663 0.09624346 0.16669542 0.04091914 0.06930294 0.15184516 0.04692222 -0.00866132 -0.00038946 -0.02592205 0.07461740 0.04047858 -0.03996211 -0.04710636 -0.04315427 0.01841215 -0.03596711 -0.06239161
C(1) 0.00312387 0.00613715 0.03785834 5.13507299 0.27056371 0.16897035 0.36640892 0.24375797 0.10229665 -0.16185476 0.01198523 0.11687751 -0.57183271 -0.29026677 -0.21603180 0.11388887 -0.11849556 -0.28787563
C(2) 0.00932125 0.35113288 0.11228255 2.43299758 15.74699285 0.13075284 0.18665130 0.42609902 0.04648042 -0.04425265 0.02023742 0.15103735 -0.24359307 -0.66191664 -0.16270174 0.05625518 -0.06260697 -0.31151745
C(3) -0.00265990 0.02692143 0.04800887 0.29651218 0.40179884 0.59967707 0.17555020 0.19804710 0.41661998 -0.10338803 -0.01199320 -0.22864135 -0.13861298 -0.23093509 -0.37867555 0.08275911 -0.09226321 -0.14830220
A{1}(1) 0.00251256 0.01175452 0.00353350 0.15314560 0.13661393 0.02507409 0.03401969 0.46301518 0.36950808 -0.86230381 -0.26844111 -0.13293947 -0.34364746 -0.26415231 -0.23964243 0.13372512 -0.11336771 -0.48627475
A{1}(2) -0.00097316 0.00533959 -0.00048015 0.07500111 0.22958573 0.02082393 0.01159567 0.01843619 0.38968132 -0.24786760 -0.63426141 -0.06468512 -0.30198997 -0.67343234 -0.31740190 0.13754579 -0.11657030 -0.58789659
A{1}(3) -0.00054092 0.00265384 -0.00001071 0.01561010 0.01242052 0.02172550 0.00458944 0.00356300 0.00453463 -0.26183167 -0.23848233 -0.65085494 -0.13128579 -0.19747059 -0.56103349 0.12062718 -0.10504330 -0.37551800
A{2}(1) -0.00236299 -0.00446728 -0.00163927 -0.05680914 -0.02719929 -0.01240076 -0.02463457 -0.00521284 -0.00273094 0.02399048 0.25603773 0.22814431 -0.14026843 0.04329838 0.03773887 -0.04477207 0.02294058 0.27059897
A{2}(2) 0.00066000 0.00373932 0.00223970 0.00199667 0.00590392 -0.00068278 -0.00364000 -0.00633127 -0.00118063 0.00291548 0.00540473 0.22511072 -0.05742195 -0.08972318 -0.01474355 0.02368185 -0.05593127 0.30116886
A{2}(3) 0.00049706 0.00032820 0.00084766 0.01358437 0.03074104 -0.00908132 -0.00125763 -0.00045048 -0.00224797 0.00181244 0.00084883 0.00263070 -0.19121212 -0.13637353 -0.22832903 -0.01727004 -0.00239688 0.00656221
B(1) 0.00003809 -0.00162970 -0.00058057 -0.04610888 -0.03439585 -0.00381949 -0.00225538 -0.00145905 -0.00031458 -0.00077308 -0.00015021 -0.00034897 0.00126615 0.41034424 0.35024732 -0.17485173 0.18624660 0.21815160
B(2) 0.00025179 -0.00286878 -0.00076757 -0.02625118 -0.10482872 -0.00713718 -0.00194445 -0.00364928 -0.00053070 0.00026765 -0.00026325 -0.00027915 0.00058273 0.00159278 0.37203441 -0.17991886 0.19085234 0.28927993
B(3) 0.00010198 -0.00100730 -0.00040185 -0.01116524 -0.01472544 -0.00668811 -0.00100811 -0.00098293 -0.00086166 0.00013332 -0.00002472 -0.00026710 0.00028425 0.00033864 0.00052018 -0.12138170 0.13038243 0.28208230
DCC(A) -0.00088578 0.00302071 0.00020357 0.00698873 0.00604512 0.00173547 0.00066792 0.00050574 0.00021997 -0.00018779 0.00004715 -0.00002399 -0.00016848 -0.00019445 -0.00007497 0.00073331 -0.96546018 -0.09399998
DCC(B) 0.00074366 -0.00374829 -0.00048101 -0.00879551 -0.00813780 -0.00234031 -0.00068492 -0.00051845 -0.00023170 0.00011639 -0.00013469 -0.00000403 0.00021708 0.00024949 0.00009741 -0.00085637 0.00107293 0.10070323
Shape -0.00399337 -0.01699480 -0.00622985 -0.15953944 -0.30232290 -0.02808641 -0.02193495 -0.01952209 -0.00618431 0.01025027 0.00541486 0.00008231 0.00189841 0.00282349 0.00157341 -0.00062253 0.00080671 0.05981081

First of all, the bottom of the output (produced by the VCV option on the GARCH) isn't really helpful, so ignore it. (And get rid of the VCV option so you don't waste space with it).

I would say that you have serious problems either with your model or with the data. The 2.89 shape parameter is about as extreme as that can get, and indicates that you have some off-the-scale outliers. I'm not sure what your data are, but you might want to make sure that you have an adequate mean model---if you've left substantial serial correlation in the residuals, then your GARCH estimates are worthless.

What I really need is the meaning of C(1), C(2),... ,DCC(B) and the interpretation of their coefficients. I will see if there are outliers and try to handle that. Thanks.

Gabizzy wrote:What I really need is the meaning of C(1), C(2),... ,DCC(B) and the interpretation of their coefficients. I will see if there are outliers and try to handle that. Thanks.

The C's are the variance intercepts in the three univariate GARCH models. DCC(A) and DCC(B) are the A and B coefficients in the GARCH model that is used to generate the time-varying correlations.

Thanks. What about the A{1}(1), A{1}(2),... B(3)? Need to know the meaning of all.

By default, you're using VARIANCES=SIMPLE (with two "A" lags). See the description at

https://estima.com/ratshelp/garchmvrpf. ... _Output_CC

Please can I get the line code for forecasting using DCC Garch (1,2) on Rats

See https://estima.com/forum/viewtopic.php?f=11&t=2104