bdp_dw_est_post_stats: Calculating the estimated posterior mean, median and credible...

In beyondWhittle: Bayesian Spectral Inference for Time Series

Calculating the estimated posterior mean, median and credible region (tv-PSD)

Description

Calculating the estimated posterior mean, median and credible region (tv-PSD)

Usage

bdp_dw_est_post_stats(post_sample, rescaled_time, freq, unif_CR = FALSE)

Arguments

post_sample	the posterior sample generated by bdp_dw_mcmc.
rescaled_time, freq	numeric vectors forming a rectangular grid on which the estimated tv-PSD is evaluated.
unif_CR	a Boolean value (default FALSE) indicating whether to calculate the uniform credible region rescaled_time must be in [0,1] and freq must be in [0,\pi].

Value

list containing the following fields:

tvpsd.mean, tvpsd.median	posterior mean and pointwise posterior median (matrices of dimension length(rescaled_time) by length(freq))
tvpsd.p05, tvpsd.p95	90 percent pointwise credibility interval
tvpsd.u05, tvpsd.u95	90 percent uniform credibility interval if unif_CR = TRUE. Otherwise NA

beyondWhittle documentation built on June 22, 2024, 11:35 a.m.