The main output class for the function estimate_risk_roll()
is
portvine_roll
but in the conditional case
the child class cond_portvine_roll
with some extra slots (below visible
by the !C!) is returned.
Details
For easy access for the most important slots and some filtering functionality
have a look at the accessor methods risk_estimates()
, fitted_vines()
,
fitted_marginals()
.
Slots
risk_estimates
data.table with the columns
risk_measure
,risk_est
,alpha
,row_num
,vine_window
andrealized
(here all samples also in the conditional case are used)fitted_marginals
named list with an entry for each asset containing a
rugarch::ugarchroll
class object that encompasses the marginal model fit.fitted_vines
list of
rvinecopulib::vinecop
class objects each entry corresponds to one vine window.marginal_settings
containing the specification used for the ARMA-GARCH fitting i.e. marginal models. Is of class
marginal_settings
.vine_settings
containing the specifications used for the vine fitting. Is of class
vine_settings
.risk_measures
a character vector displaying the estimated risk measures.
alpha
numeric vector in (0,1) displaying the confidence levels used when estimating the risk measures.
weights
the numeric positive weights of the assets. (Matrix with each row corresponding to one vine window) The weights of conditional variables are always 0.
cond_estimation
logical value indicating whether the conditional estimation approach for the risk measures was used.
n_samples
positive numeric count displaying how many return samples were used for the risk measure estimation.
time_taken
numeric value displaying how many minutes the whole estimation process took.
cond_risk_estimates
!C! data.table with the same columns as the
risk_estimate
slot has + the additional conditional columns with the respective conditioning value and the column charactercond_u
that indicates the used conditional quantile level or the conditional value corresponding to the residual one time unit prior with "prior_resid" or the realized residual with "resid".cond_vars
!C! character vector with the names of the variables that were used to sample conditionally from.
cond_u
!C! a numeric vector specifying the corresponding quantiles in (0,1) of the conditional variable(s) conditioned on which the conditional risk measures were calculated.