fc_fit: Fit a factor copula model

In bonartm/factorcopula: High Dimensional Depence Modelling with Factor Copulas

Description

Fit a factor copula model

Usage

fc_fit(Y, factor, error, beta, lower, upper, k = rep(1, ncol(Y)), S = 25 *
  nrow(Y), se = FALSE, B = 1000, control.first.stage = list(algorithm =
  "NLOPT_GN_DIRECT_L", stopval = 0, xtol_rel = 1e-04, maxeval = 200),
  control.second.stage = list(algorithm = "NLOPT_LN_SBPLX", stopval = 0,
  xtol_rel = 1e-16, maxeval = 10000))

Arguments

Y	A dataframe or matrix like object
factor	specification of latent variables, see config_factor
error	specification of error term, see config_error
beta	specification of parameter matrix, see config_beta
lower	Lower bound for optimazation: Named vector with parameters
upper	Upper bound for optimazation: Named vector with parameters
k	A vector with length ncol(Y) defining the groups (e.q. equi-dependence or block-euqidependence model)
S	The number of simulations to use
se	Wether to estimate standard errors and confidence intervalls
B	number of bootstrap replication for estimating standard error
control.first.stage	named list of options passed to the nloptr method. A full description of all options is shown by the function nloptr::nloptr.print.options().
control.second.stage	named list of options passed to the nloptr method. A full description of all options is shown by the function nloptr::nloptr.print.options().

Value

a list of optimization results

bonartm/factorcopula documentation built on April 19, 2020, 9:17 p.m.