Sei sulla pagina 1di 3

* Testing As mentioned by Tse (2002), diagnostics for conditional

heteroscedasticity models applied in the literature can be divided into three


categories: portmanteau tests of the Box�Pierce�Ljung type, residual-based
diagnostics and Lagrange multiplier tests

Checking the overall adequacy of a model Diebold et al. (1999) (in the multivariate
case):

The most widely used diagnostics to detect ARCH effectsare probably the
Box�Pierce/Ljung�Box portmanteau tests. Following Hosking (1980), a multivariate
version of the Ljung�Box test statistic:
wntstmvq
mvreg

Residual-Based Diagnostics: An advantage of the residual-based diagnostics is that


they focus on several distinctive aspects of possible causes of �remaining� ARCH
effects
Tse (2002) test

Lagrange multiplier tests

R STATS
archBootTest combined LM Test for Arch errors in VAR models
Performs the bootstrap combined Lagrange multiplier (LM) test for autoregressive
conditional heteroskedastic (ARCH) errors in vector autoregressive (VAR) models of
Catani and Ahlgren (2016)

VARtests
var.2c <- VAR(Canada, p = 2, type = "const")
arch.test(var.2c)
arch.test(x, lags.single = 16, lags.multi = 5, multivariate.only = TRUE)

fit <- VARfit(y = VodafoneCDS, p = 3, const = TRUE, trend = FALSE)


test <- archBootTest(fit = fit, h = 5, B = 199, CA = TRUE, ET = TRUE, MARCH = TRUE,
dist = "norm")

> test<-archBootTest(fit = fit, h = 4, B = 199, CA = TRUE, ET = TRUE, MARCH = TRUE,


dist = "norm")

Estimated time to complete the 199 bootstrap simulations: 2.5 mins


Running Bootstrap: 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Test for ARCH errors in VAR models:

h: 4
B: 199
Error distribution: norm

-------------------- Multivariate tests: ---------------------


LM df Asy.PV Boot.PV
CA 1 NA NA 0.005 **
ET 53965 24 0 0.005 **
MARCH 11671 1764 0 0.005 **
---
Signif. codes: 0 �***� 0.001 �**� 0.01 �*� 0.05 �.� 0.1 � � 1
--------------------------------------------------------------

---------- Catani and Ahlgren equation by equation: ----------


LMi df Asy.PV Boot.PV
beta_0 293.20 4.00 0 0.005 **
beta_1 156.63 4.00 0 0.005 **
beta_2 278.44 4.00 0 0.005 **
beta_3 214.73 4.00 0 0.005 **
tau1 206.61 4.00 0 0.005 **
tau2 406.90 4.00 0 0.005 **
---
Signif. codes: 0 �***� 0.001 �**� 0.01 �*� 0.05 �.� 0.1 � � 1
--------------------------------------------------------------

Test time (secs): 186.9175

Wild Bootstrap simulations started at 2019-05-16 18:42:44

Estimated time to complete the 199 bootstrap simulations: 4.4 mins


Running Bootstrap: 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

------ Multivariate test for error autocorrelations (AC) ------

h: 4 B: 199

Recursive Wild Bootstrap:

Q df Asy.PV WB.PV
LM 761.42 144.00 0.0000 0.010 **
HC3 140.13 144.00 0.5756 0.195
---
Signif. codes: 0 �***� 0.001 �**� 0.01 �*� 0.05 �.� 0.1 � � 1
--------------------------------------------------------------

0 of the bootstrap simulations failed and had to be resimulated


Test time: 9.9 mins>

> wb <- wildBoot(test = test2, WBtype = c("recursive", "fixed"), B = 199, WBdist =


"rademacher", HCtype = c("HC1","HC2","HC0", "HC3"))

Wild Bootstrap simulations started at 2019-05-16 21:58:32

Estimated time to complete the 199 bootstrap simulations: 20 mins


Running Bootstrap: 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

------ Multivariate test for error autocorrelations (AC) ------

h: 4 B: 199

Recursive Wild Bootstrap:

Q df Asy.PV WB.PV
HC0 151.66 144.00 0.3146 0.195
HC1 151.29 144.00 0.3221 0.195
HC2 145.58 144.00 0.4475 0.210
HC3 140.13 144.00 0.5756 0.230

Fixed Wild Bootstrap:

Q df Asy.PV WB.PV
HC0 151.66 144.00 0.3146 0.215
HC1 151.29 144.00 0.3221 0.215
HC2 145.58 144.00 0.4475 0.200
HC3 140.13 144.00 0.5756 0.175
--------------------------------------------------------------

0 of the bootstrap simulations failed and had to be resimulated


Test time: 9.8 hours>

Potrebbero piacerti anche