Estimate GARCH Parameters for Multivariate Models (R Helper)

Two-stage weighted estimation for multivariate GARCH models: Stage 1: Univariate GARCH parameters for each series Stage 2: Multivariate dependence parameters (DCC/Copula) or decomposition (GOGARCH)

Usage

estimate_garch_weighted_r(
  residuals,
  weights,
  spec,
  model_type = "univariate",
  diagnostics = NULL,
  iteration = NULL,
  state = NULL,
  verbose = FALSE,
  dcc_threshold = 0.02,
  dcc_criterion = "bic",
  force_constant = FALSE
)

Arguments

residuals: Matrix of residuals (T x k) for multivariate case
weights: Vector of weights from E-step
spec: Model specification
model_type: Either "univariate" or "multivariate"
diagnostics: diagnostics
iteration: iteration
state: state
verbose: verbose
dcc_threshold: dcc_threshold
dcc_criterion: dcc_criterion
force_constant: force_constant

Value

List with coefficients and warnings