Bootstrap Standard Errors for DCC(1,1) Parameters

Compute bootstrap standard errors using either parametric or residual resampling.

Usage

dcc11_bootstrap_se(
  std_resid,
  weights,
  Qbar,
  mle_params,
  n_boot = 200,
  method = c("residual", "parametric"),
  distribution = "mvn",
  shape = NULL,
  verbose = TRUE,
  seed = NULL
)

Arguments

std_resid

T x k matrix of standardized residuals

weights

T-vector of observation weights

Qbar

k x k unconditional correlation matrix

mle_params

MLE estimates c(alpha, beta) or c(alpha, beta, shape)

n_boot

Number of bootstrap replications (default 200)

method

Bootstrap method: "parametric" or "residual"

distribution

"mvn" or "mvt"

shape

Shape parameter for MVT (required if distribution="mvt")

verbose

Print progress

seed

Random seed for reproducibility

Value

List with bootstrap results