0% found this document useful (0 votes)
78 views28 pages

Arch Garch

The document provides the results of various lag order selection criteria tests for autoregressive (AR) and autoregressive conditional heteroskedasticity (ARCH) models. Based on the Akaike information criterion and Schwarz criterion, an AR(7) model is selected for the endogenous variable RAPPLE. For the endogenous variable RESID2, an ARCH(5) model is indicated as optimal according to the Akaike information criterion and Schwarz criterion. Finally, a GARCH(1,1) model is selected for RAPPLE based on the Akaike information criterion, Schwarz criterion, and Hannan-Quinn criterion.

Uploaded by

sergio
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as XLSX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
78 views28 pages

Arch Garch

The document provides the results of various lag order selection criteria tests for autoregressive (AR) and autoregressive conditional heteroskedasticity (ARCH) models. Based on the Akaike information criterion and Schwarz criterion, an AR(7) model is selected for the endogenous variable RAPPLE. For the endogenous variable RESID2, an ARCH(5) model is indicated as optimal according to the Akaike information criterion and Schwarz criterion. Finally, a GARCH(1,1) model is selected for RAPPLE based on the Akaike information criterion, Schwarz criterion, and Hannan-Quinn criterion.

Uploaded by

sergio
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as XLSX, PDF, TXT or read online on Scribd
You are on page 1/ 28

VAR Lag Order Selection Criteria

Endogenous variables: RAPPLE


Exogenous variables: C
Date: 06/12/19 Time: 17:49
Sample: 5/22/2014 5/21/2019
Included observations: 1249

Lag LogL LR FPE AIC SC HQ

0 3437.338 NA 0.000239* -5.502543* -5.498436* -5.500999*


1 3437.406 0.135734 0.000239 -5.501051 -5.492836 -5.497963
2 3438.354 1.890472 0.000239 -5.500967 -5.488645 -5.496334
3 3438.59 0.469915 0.000239 -5.499743 -5.483314 -5.493566
4 3438.616 0.051911 0.00024 -5.498183 -5.477647 -5.490463
5 3438.62 0.008131 0.00024 -5.496589 -5.471945 -5.487324
6 3438.62 4.03E-06 0.00024 -5.494987 -5.466236 -5.484178
7 3443.425 9.548334* 0.000239 -5.50108 -5.468221 -5.488727
8 3444.183 1.506606 0.000239 -5.500694 -5.463728 -5.486797

* indicates lag order selected by the criterion


LR: sequential modified LR test statistic (each test at 5% level)
FPE: Final prediction error
AIC: Akaike information criterion
SC: Schwarz information criterion
HQ: Hannan-Quinn information criterion
Criterio AR(1) AR(2) AR(3) AR(4) AR(5) AR(6)
Akaike info criterion -5.503277 -5.504938 -5.503050 -5.502933 -5.502189 -5.501569
Schwarz criterion -5.495099 -5.496755 -5.494862 -5.494740 -5.493990 -5.493365
Hannan-Quinn criterion -5.500203 -5.501862 -5.499972 -5.499853 -5.499107 -5.498485

Dependent Variable: RAPPLE

Variable Coefficien... Std. Error

C 0.000755 0.000478
AR(7) 0.088654 0.028401

R-squared 0.007747 Mean depend


Adjusted R-squared 0.006952 S.D. depend
S.E. of regression 0.015386 Akaike info c
Sum squared resid 0.295431 Schwarz crit
Log likelihood 3445.213 Hannan-Quin
F-statistic 9.743930 Durbin-Wats
Prob(F-statistic) 0.001840

Inverted AR Roots .71 .44+.55...


Ningún criterio indican que es un modelo AR (n) -.16+.69i -.64+.31... -

Heteroskedasticity Test: ARCH

F-statistic 8.206308 Prob. F(8,123


Obs*R-squared 62.78662 Prob. Chi-Squ
AR(7) AR(8)
-5.509141* -5.502015
-5.500931* -5.493801
-5.506054* -5.498927

PLE

efficien... Std. Error t-Statistic Prob.

0.000755 0.000478 1.581518 0.1140


0.088654 0.028401 3.121527 0.0018

0.007747 Mean dependent var 0.000763


0.006952 S.D. dependent var 0.015440
0.015386 Akaike info criterion -5.509141
0.295431 Schwarz criterion -5.500931
3445.213 Hannan-Quinn criter. -5.506054
9.743930 Durbin-Watson stat 1.972282
0.001840

.71 .44+.55... .44-.55i -.16-.69i


6+.69i -.64+.31... -.64-.31i

RCH

8.206308 Prob. F(8,1233) 0.0000


62.78662 Prob. Chi-Square(8) 0.0000
VAR Lag Order Selection Criteria
Endogenous variables: RESID2
Exogenous variables: C
Date: 06/13/19 Time: 13:50
Sample: 5/22/2014 5/21/2019
Included observations: 1247

Lag LogL LR FPE AIC SC HQ

0 7531.926 NA 3.33E-07 -12.07847 -12.07436 -12.07692


1 7539.842 15.80753 3.29E-07 -12.08956 -12.08134 -12.08647
2 7540.62 1.551244 3.29E-07 -12.08921 -12.07687 -12.08457
3 7545.2 9.131373 3.27E-07 -12.09495 -12.0785 -12.08876
4 7545.462 0.522251 3.28E-07 -12.09376 -12.0732 -12.08603
5 7558.245 25.44177 3.21E-07 -12.11266 -12.08799* -12.10338*
6 7559.809 3.11E+00 3.21E-07 -12.11357 -12.08478 -12.10274
7 7560.954 2.275201 3.21E-07 -12.1138 -12.0809 -12.10143
8 7563.703 5.457271* 3.20e-07* -12.11660* -12.07959 -12.10269
9 7564.092 0.773054 3.20E-07 -12.11563 -12.0745 -12.10016
10 7564.407 0.622639 3.21E-07 -12.11453 -12.06929 -12.09752

* indicates lag order selected by the criterion


LR: sequential modified LR test statistic (each test at 5% level)
FPE: Final prediction error
AIC: Akaike information criterion
SC: Schwarz information criterion
HQ: Hannan-Quinn information criterion
0 1 2 3 4
Criterio Sin Lag ARCH(1) ARCH(2) ARCH(3) ARCH(4)
Akaike info criterion -12.07847 -12.08956 -12.08921 -12.09495 -12.09376
Schwarz criterion -12.07436 -12.08134 -12.07687 -12.0785 -12.0732
Hannan-Quinn criterion -12.07692 -12.08647 -12.08457 -12.08876 -12.08603

Orden del ARCH

Ningún criterio indican que es un modelo AR (n)

Variable Coefficien... Std. Error z-Statistic Prob.

C 0.001297 0.000389 3.335885 0.0009


AR(7) 0.062122 0.025413 2.444467 0.0145

Variance Equation

C 0.000120 6.48E-06 18.46153 0.0000


RESID(-1)^2 0.157522 0.033522 4.699057 0.0000
RESID(-2)^2 0.051604 0.023921 2.157255 0.0310
RESID(-3)^2 0.137837 0.024713 5.577425 0.0000
RESID(-4)^2 0.065214 0.025630 2.544407 0.0109
RESID(-5)^2 0.116274 0.019351 6.008783 0.0000

Modelo ARCH(5)

RESID(-1)^2 0.157522
RESID(-2)^2 0.051604
RESID(-3)^2 0.137837
RESID(-4)^2 0.065214
RESID(-5)^2 0.116274
Suma 0.528451
5 6 7 8 9 10
ARCH(5) ARCH(6) Criterio AR(1) AR(2) AR(3)
-12.11266 -12.11357 Akaike info criterion -12.09619 -12.08496 -12.0908
-12.08799* -12.08478 Schwarz criterion -12.08801 -12.07677 -12.08261
-12.10338* -12.10274 Hannan-Quinn criterion -12.09312 -12.08188 -12.08772

en del ARCH
AR(4) AR(5) AR(6)
-12.08288 -12.10402* -12.0875
-12.07468 -12.09582* -12.0793
-12.0798 -12.10094* -12.08442
Criterio GARH(1,1) GARH(2,1) GARH(1,2) GARH(2,2) GARH(1,3)
Akaike info criterion -5.603095 -5.603095 -5.604194 -5.60313 -5.605184*
Schwarz criterion -5.58265* -5.58265 -5.579661 -5.574508 -5.576561
Hannan-Quinn criterion -5.595411* -5.595411 -5.594974 -5.592373 -5.594426

Dependent Variable: RAPPLE

Variable Coefficien... Std. Error z-Statistic Prob.

C 0.001500 0.000381 3.940917 0.0001


AR(1) 0.007540 0.031942 0.236056 0.8134

Variance Equation

C 1.91E-05 3.36E-06 5.692213 0.0000


RESID(-1)^2 0.112736 0.021277 5.298532 0.0000
GARCH(-1) 0.810427 0.030087 26.93611 0.0000
Modelo GARCH (1,1)

ht = 0.0000191031+ 0.112736 Ԑ2 t-1+


0.810427 ht-1

α0= 0.000019 C
RESID(-1)^2
α1= 0.112736 El shock de las noticias GARCH(-1)

β1= 0.810427 Volatilidad de un periodo anterior

α1 + β1 = 0.923163

El modelo GARCH es estacionario


Varianza
Rezagada
∑24_(𝑗=1)^𝑝▒
∑24_(𝑖=1)^𝑚▒ 〖 "
〖�𝑗ℎ𝑡−𝑖〗 α" 𝑖�2𝑡−𝑖 〗 Noticias o
Shocks

∑=

p
j
itjh
1
β
GARH(2,3) GARH(3,3)
-5.604235 -5.603817
-5.571524 -5.567016
-5.591941 -5.589986

Modelo GARCH (1,1)


Heteroskedasticity Test: ARCH

F-statistic 0.034
Obs*R-squared 0.034

C 1.9E-05 3.36E-06 5.692213 0.0000


RESID(-1)^2 0.112736 0.021277 5.298532 0.0000
GARCH(-1) 0.810427 0.030087 26.93611 0.0000

∑ = βjht-i

p
j
itjh
1
β
Heteroskedasticity Test: ARCH

F-statistic 0.034937 Prob. F(1,1253) 0.8518


Obs*R-squared 0.034992 Prob. Chi-Square(1) 0.8516
VAR Lag Order Selection Criteria
Endogenous variables: RSP500
Exogenous variables: C
Date: 06/14/19 Time: 03:07
Sample: 5/22/2014 5/21/2019
Included observations: 1247

Lag LogL LR FPE AIC SC HQ

0 4188.97 NA* 7.09e-05* -6.716872* -6.712760* -6.715326*


1 4189.228 0.515048 7.09E-05 -6.715682 -6.707457 -6.71259
2 4189.963 1.466062 7.10E-05 -6.715257 -6.702919 -6.710618
3 4190.108 0.290665 7.11E-05 -6.713887 -6.697436 -6.707702
4 4191.402 2.577693 7.10E-05 -6.714358 -6.693795 -6.706627
5 4191.9 0.989811 7.11E-05 -6.713552 -6.688876 -6.704274
6 4191.98 1.60E-01 7.12E-05 -6.712077 -6.683288 -6.701253
7 4192.256 0.549392 7.13E-05 -6.710916 -6.678015 -6.698546
8 4193.934 3.331496 7.12E-05 -6.712004 -6.67499 -6.698087
9 4194.612 1.344397 7.12E-05 -6.711487 -6.67036 -6.696024
10 4194.833 0.438229 7.13E-05 -6.710237 -6.664998 -6.693228

* indicates lag order selected by the criterion


LR: sequential modified LR test statistic (each test at 5% level)
FPE: Final prediction error
AIC: Akaike information criterion
SC: Schwarz information criterion
HQ: Hannan-Quinn information criterion
1 2 3 4 5 6
Criterio AR(1) AR(2) AR(3) AR(4) AR(5) AR(6)
Akaike info criterion -6.715682 -6.715257 -6.713887 -6.714358 -6.713552 -6.712077
Schwarz criterion -6.707457 -6.702919 -6.697436 -6.693795 -6.688876 -6.683288
Hannan-Quinn criterion -6.71259 -6.710618 -6.707702 -6.706627 -6.704274 -6.701253

Orden del ARCH Heteroskedasticity Test: ARCH

F-statistic 27.43949
Obs*R-squared 187.8537

Ningún criterio indican que es un modelo AR (n)

Modelo ARCH(5)

RESID(-1)^2
RESID(-2)^2
RESID(-3)^2
RESID(-4)^2
RESID(-5)^2
Suma
7 8 9 10
Criterio AR(1) AR(2) AR(3) AR(4) AR(5)
Akaike info criterion -12.09619 -12.08496 -12.0908 -12.08288 -12.10402*
Schwarz criterion -12.08801 -12.07677 -12.08261 -12.07468 -12.09582*
Hannan-Quinn criterion -12.09312 -12.08188 -12.08772 -12.0798 -12.10094*

Heteroskedasticity Test: ARCH

F-statistic 27.43949 Prob. F(8,1240) 0.0000


Obs*R-squared 187.8537 Prob. Chi-Square(8) 0.0000
AR(6)
-12.0875
-12.0793
-12.08442
VAR Lag Order Selection Criteria
Endogenous variables: RESID2SP500
Exogenous variables: C
Date: 06/14/19 Time: 04:02
Sample: 5/22/2014 5/21/2019
Included observations: 1247

Lag LogL LR FPE AIC SC HQ

0 9051.325 NA 2.91E-08 -14.51536 -14.51124 -14.51381


1 9105.519 108.2134 2.67E-08 -14.60067 -14.59245 -14.59758
2 9119.381 27.6573 2.62E-08 -14.6213 -14.60896 -14.61666
3 9128.846 18.8702 2.58E-08 -14.63488 -14.61843 -14.62869
4 9136.091 14.43019 2.55E-08 -14.64489 -14.62433 -14.63716
5 9136.091 0.001336 2.56E-08 -14.64329 -14.61861 -14.63401
6 9152.146 31.93029* 2.50e-08* -14.66744* -14.63865* -14.65661*
7 9152.838 1.374824 2.50E-08 -14.66694 -14.63404 -14.65457
8 9152.899 0.120272 2.50E-08 -14.66544 -14.62842 -14.65152
9 9152.988 0.177602 2.51E-08 -14.66397 -14.62285 -14.64851
10 9154.683 3.358356 2.50E-08 -14.66509 -14.61985 -14.64808

* indicates lag order selected by the criterion


LR: sequential modified LR test statistic (each test at 5% level)
FPE: Final prediction error
AIC: Akaike information criterion
SC: Schwarz information criterion
HQ: Hannan-Quinn information criterion
0 1 2 3 4
Criterio Sin Lag ARCH(1) ARCH(2) ARCH(3) ARCH(4)
Akaike info criterion -6.855753 -6.909083 -6.948987 -6.990654
Schwarz criterion -6.843494 -6.892738 -6.928555 -6.966136
Hannan-Quinn criterion -6.851145 -6.90294 -6.941308 -6.98144

Orden del ARCH

Ningún criterio indican que es un modelo AR (n)

Variable Coefficien... Std. Error z-Statistic Prob.

C 0.000782 0.000180 4.342800 0.0000

Variance Equation

C 1.74E-05 1.28E-06 13.64531 0.0000


RESID(-1)^2 0.210919 0.025181 8.376112 0.0000
RESID(-2)^2 0.120575 0.034003 3.546061 0.0004
RESID(-3)^2 0.147168 0.028871 5.097482 0.0000
RESID(-4)^2 0.178434 0.029688 6.010317 0.0000
RESID(-5)^2 0.046602 0.017905 2.602710 0.0092
RESID(-6)^2 0.097454 0.029304 3.325582 0.0009

Modelo ARCH(5)

RESID(-1)^2 0.210919
RESID(-2)^2 0.120575
RESID(-3)^2 0.147168
RESID(-4)^2 0.178434
RESID(-5)^2 0.046602
RESID(-6)^2 0.097454
Suma 0.801152
5 6 7 8 9 10
ARCH(5) ARCH(6) ARCH(7) Criterio AR(1) AR(2)
-7.004798 -7.012767* -7.011263 Akaike info criterion -6.855753 -6.909083
-6.976193 -6.980077* -6.974486 Schwarz criterion -6.843494 -6.892738
-6.994047 -7.000481* -6.997441 Hannan-Quinn criterion -6.851145 -6.90294

n del ARCH
AR(3) AR(4) AR(5) AR(6) AR(7)
-6.948987 -6.990654 -7.004798 -7.012767 -7.011263
-6.928555 -6.966136 -6.976193 -6.980077 -6.974486
-6.941308 -6.98144 -6.994047 -7.000481 -6.997441
Criterio GARH(1,1) GARH(2,1) GARH(1,2) GARH(2,2) GARH(1,3)
Akaike info criterion 8.777848 8.652079* 11.99615 11.53945 11.53095
Schwarz criterion 8.798279 8.676597* 12.02067 11.56806 11.55956
Hannan-Quinn criterion 8.785526 8.661294* 12.00537 11.5502 11.5417

Dependent Variable: SP500

Variable Coefficien... Std. Error z-Statistic Prob.

C 18818.93 271198.9 0.069392 0.9447


AR(1) 0.999917 0.001362 733.9348 0.0000

Variance Equation

C 484.1437 23.44118 20.65356 0.0000


RESID(-1)^2 0.338887 0.037635 9.004621 0.0000
RESID(-2)^2 0.316727 0.037764 8.387110 0.0000
GARCH(-1) -0.926373 0.067234 -13.77834 0.0000

ht = 0.0000191031+ 0.112736 Ԑ2 t-1+


0.810427 ht-1

α0= 484.143700

α1= 0.338887 El shock de las noticias

α2= 0.316727

β1= -0.926373 Volatilidad de un periodo anterior


α1 + β1 = -0.270759

El modelo GARCH tiene coeficiente negativo


Varianza
Rezagada
∑24_(𝑗=1)^𝑝▒
∑24_(𝑖=1)^𝑚▒ 〖 "
〖�𝑗ℎ𝑡−𝑖〗 α" 𝑖�2𝑡−𝑖 〗 Noticias o
Shocks

∑=

p
j
itjh
1
β
GARH(2,3) GARH(3,3)
11.63717 11.57498
11.66986 11.61176
11.64946 11.5888

Date: 06/14/19 Time: 04:56


Sample: 5/22/2014 5/21/2019
Included observations: 1257
Q-statistic probabilities adjusted for 1 ARMA term

Autocorrelation Partial Correlation AC PA... Q-St... Pro...

1-0.01... -0.01... 0.146...


Modelo GARCH (2,1) 2-0.01... -0.01... 0.612... 0.43...
3 0.02... 0.02... 1.643... 0.44...
4-0.01... -0.01... 1.790... 0.61...
5-0.04... -0.03... 3.819... 0.43...
6-0.01... -0.01... 4.057... 0.54...
7 0.01... 0.01... 4.503... 0.60...
8-0.05... -0.04... 7.731... 0.35...
9-0.03... -0.03... 9.517... 0.30...
1... -0.00... -0.00... 9.517... 0.39...
1... -0.00... -0.00... 9.626... 0.47...

Heteroskedasticity Test: ARCH

F-statistic 4.791838 Prob. F(2,1252) 0.0084


Obs*R-squared 9.533663 Prob. Chi-Square(2) 0.0085

Dependent Variable: SP500

Variable Coefficien... Std. Error z-Statistic Prob.

C 18818.93 271198.9 0.069392 0.9447


AR(1) 0.999917 0.001362 733.9348 0.0000

∑ = βjht-i Variance Equation



C 484.1437 23.44118 20.65356 0.0000
p RESID(-1)^2 0.338887 0.037635 9.004621 0.0000
j RESID(-2)^2 0.316727 0.037764 8.387110 0.0000
itjh GARCH(-1) -0.926373 0.067234 -13.77834 0.0000
C 484.1437 23.44118 20.65356 0.0000
RESID(-1)^2 0.338887 0.037635 9.004621 0.0000
RESID(-2)^2 0.316727 0.037764 8.387110 0.0000
GARCH(-1) -0.926373 0.067234 -13.77834 0.0000
1
β
. Q-St... Pro...

.. 0.146...
.. 0.612... 0.43...
.. 1.643... 0.44...
.. 1.790... 0.61...
.. 3.819... 0.43...
.. 4.057... 0.54...
.. 4.503... 0.60...
.. 7.731... 0.35...
.. 9.517... 0.30...
.. 9.517... 0.39...
.. 9.626... 0.47...

0.0084
(2) 0.0085

-Statistic Prob.

0.069392 0.9447
733.9348 0.0000

20.65356 0.0000
9.004621 0.0000
8.387110 0.0000
13.77834 0.0000
20.65356 0.0000
9.004621 0.0000
8.387110 0.0000
13.77834 0.0000

You might also like