Method for creating a Copula-GARCH specification object prior to fitting.
cgarchspec(uspec, VAR = FALSE, robust = FALSE, lag = 1, lag.max = NULL,
lag.criterion = c("AIC", "HQ", "SC", "FPE"), external.regressors = NULL,
robust.control = list(gamma = 0.25, delta = 0.01, nc = 10, ns = 500),
dccOrder = c(1, 1), asymmetric = FALSE,
distribution.model = list(copula = c("mvnorm", "mvt"),
method = c("Kendall", "ML"), time.varying = FALSE,
transformation = c("parametric", "empirical", "spd")),
start.pars = list(), fixed.pars = list())A '>uGARCHmultispec object created by calling
multispec on a list of univariate GARCH specifications.
Whether to fit a VAR model for the conditional mean.
Whether to use the robust version of VAR.
The VAR lag.
The maximum VAR lag to search for best fit.
The criterion to use for choosing the best lag when lag.max is not NULL.
Allows for a matrix of common pre-lagged external regressors for the VAR option.
The tuning parameters to the robust regression including the proportion to trim (“gamma”), the critical value for reweighted estimator (“delta”), the number of subsets (“ns”) and the number of C-steps (“nc”.
The DCC autoregressive order.
Whether to include an asymmetry term to the DCC model (thus estimating the aDCC).
The Copula distribution model. Currently the multivariate Normal and Student Copula are supported.
Whether to fit a dynamic DCC Copula.
The type of transformation to apply to the marginal innovations of the GARCH fitted models. Supported methods are parametric (Inference Function of Margins), empirical (Pseudo ML), and Semi-Parametric using a kernel interior and GPD tails (via the ‘spd’ package).
(optional) Starting values for the DCC parameters (starting values for the univariate garch specification should be passed directly via the ‘uspec’ object).
(optional) Fixed DCC parameters.
A '>cGARCHspec object containing details of the Copula-GARCH
specification.
The transformation method allows for parametric (Inference-Functions for Margins),
empirical (Pseudo-Likelihood) and semi-parametric (via the spd package).
When the Student Copula is jointly estimated with student margins having so that
a common shape parameter is obtained, this results in the multivariate Student
distribution. When estimating the Student Copula with disparate margins, a
meta-student distribution is obtained. Additionally, the correlation parameter
in the static Student Copula may be estimated either by Kendall's tau
transformation or Maximum Likelihood.
The robust option allows for a robust version of VAR based on the
multivariate Least Trimmed Squares Estimator described in Croux and Joossens
(2008).