Description Usage Arguments Details Value
View source: R/LG_boot_approx.R
This is a wrapper for the function LG_approx
, to be used
when we want to use bootstrapbased statistics in our analysis of
e.g. its local Gaussian spectra.
1 2 3  LG_boot_approx(save_dir = NULL, TS_boot, lag_max, LG_points,
.bws_mixture, bw_points, .bws_fixed, .bws_fixed_only, content_details,
LG_type)

save_dir 
A specification of the directory to be used when
saving (and loading) data. The default value 
TS_boot 
The matrix of bootstrapreplicates produced by

lag_max 
The number of lags to include in the analysis. 
LG_points 
An array that specifies the point at which it is
desired to compute the local Gaussian estimates. The default
value 
.bws_mixture 
An argument that specifies how the global
bandwidths and those obtained by the nearestneighbour strategy
should be combined. The three available options are

bw_points 
A vector, default 
.bws_fixed 
A vector of nonnegative real values, that can be
used to specify fixed values for the bandwidths (which might be
of interest to do in a preliminary analysis). The default
value 
.bws_fixed_only 
A logic value, default 
content_details 
A character string from 
LG_type 
One of 
This function can be called manually from the workspace,
but the intention is that it only should be called from
LG_boot_approx_scribe
, since that will ensure that the
arguments are properly recorded and that the result are saved
to appropriately named files. In order to dissuade users from
calling this (often quite time consuming) function directly, no
default values have been specified for the arguments.
data.0, a dataframe with a column levels
, which for
each value is matched with bw_points
(the prescribed
numbers of observations), bw
the corresponding local
bandwidths, and then for each replicate in TS_boot
the
estimated local Gaussian approximation parameters mu
and
sig
obtained when loclik2
has been used on the
specified level
. The estimated value of the density at
the point of interest, f.est
are also included. (Note:
The square of sig
gives the value we will need later on
in the quest for the local Gaussian spectra based on local
Gaussian autocorrelations.)
data.h, a dataframe with columns lag
and
levels
that specify the points of interest, and for each
combination of these we have information about bw_points
(the prescribed numbers of observations), bw
the
corresponding local bandwidths, and then for each replicate in
TS_boot
we have from localgauss
the coefficients
mu_1
, mu_2
, sig_1
, sig_2
and
rho
for the local bivariate Gaussian approximation. The
column par_one
gives the extracted local Gaussian
autocovariance that will be used in the computation of the
local Gaussian spectra, and f.est
gives the estimated
density.
eflag, vector
, which gives the sum of the exit flags
from localgauss
. If this is something else than 0, then
it means that some value(s) for that bootstrapreplicate not
should be trusted. (To find the actual problematic values,
pick out the relevant replicate(s) and create a
TS.object
based on it, then run LG_approx
on that
object.)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.