computeCIs-QPBoot: Pointwise Confidence Intervalls

Description Usage Arguments Value

Description

Depending on method this calculates pointwise confidence intervalls for a smoothed periodgram that belongs to a time series defined by model and param. If (method = "quantiles") it computes the α/2 and 1-α/2 quantiles from the Values of the simulated smoothed periodograms and returns those. If (method = "norm") it uses the asymptotic normality of the smoothed periodograms by estimating mean and standard deviation for each frequency and computing the α/2 and 1-α/2 quantiles from a normal distribution with the estimated parameters.

Usage

1
2
computeCIs(object, alpha = 0.05, method = c("quantiles", "norm"),
  levels = object@sPG@levels[[1]])

Arguments

object

the QPBoot object that will be plotted

alpha

the significiant level of the confidence intervalls, defaults to 0.05

method

either "quantile" or "norm", determines how the confidence intervalls are calculated. see description for details

levels

numeric vector containing values between 0 and 1 for which the smoothedPG. Will be estimated. These are the quantiles levels that are used for the validation

Value

Returns a list with four elements

q_up
q_low
mean
sd

stefanbirr/QPBoot documentation built on May 30, 2019, 10:43 a.m.