LG_normalisation_adjustment: Create a normalised time series from the original ones.

Description Usage Arguments Details Value

View source: R/LG_normalisation_adjustment.R

Description

Create a normalised time series from the original ones.

Usage

1
LG_normalisation_adjustment(TS, .adjustment_rule = 0, .remove_ties = TRUE)

Arguments

TS

The time series object to normalise. It's assumed that this is structured in an array, with one dimension-name being "content".

.adjustment_rule

A non-negative number that will be stored as an attribute that later on can be used in order to add some additional scaling (in addition to the one from the windows function) when the estimated local Gaussian auto- and cross-correlations are used for the computation of the local Gaussian spectra. The default value 0 ensures that only the scaling from the window-function is used.

.remove_ties

A logical value, default TRUE, in which case the presence of ties will trigger a minor perturbation of the data. Note: Whenever this happens, set.seed(1) will be used in order to ensure reproducibility.

Details

A normalisation of the time series under consideration is preferable when an analysis based on the local Gaussian approximations are of interest. The intention of this function is to allow for easier adjustment of the normalisation regime, i.e if the normalisation should be based on the cumulative density function (or other options that might be added later on). This function will also add some attributes to the result, with the required instruction for the finite-sample adjustments to be used when computing the local Gaussian autocorrelations.

Value

A normalised version of the time series of interest.


LAJordanger/localgaussSpec documentation built on July 28, 2017, 12:15 a.m.