bdp_dw_prior_params_gen: Generate a list of parameter values in prior elicitation

In beyondWhittle: Bayesian Spectral Inference for Time Series

Generate a list of parameter values in prior elicitation

Description

Generate a list of parameter values in prior elicitation

Usage

bdp_dw_prior_params_gen(
  M = 1,
  g0.alpha = 1,
  g0.beta = 1,
  k1.theta = 0.01,
  k2.theta = 0.01,
  tau.alpha = 0.001,
  tau.beta = 0.001,
  k1max = 100,
  k2max = 100,
  L = 20,
  bernstein1_l = 0.1,
  bernstein1_r = 0.9,
  bernstein2_l = 0.1,
  bernstein2_r = 0.9
)

Arguments

M	DP base measure constant (> 0)
g0.alpha, g0.beta	parameters of Beta base measure of DP
k1.theta	prior parameter for polynomial corresponding to rescaled time (propto exp(-k1.theta*k1*log(k1)))
k2.theta	prior parameter for polynomial corresponding to rescaled frequency (propto exp(-k2.theta*k2*log(k2)))
tau.alpha, tau.beta	prior parameters for tau (inverse gamma)
k1max	upper bound of the degrees of Bernstein polynomial corresponding to rescaled time (for pre-computation of basis functions)
k2max	upper bound of the degrees of Bernstein polynomial corresponding to rescaled frequency (for pre-computation of basis functions)
L	truncation parameter of DP in stick breaking representation
bernstein1_l, bernstein1_r	left and right truncation of Bernstein polynomial basis functions for rescaled time, 0<=bernstein1_l<bernstein1_r<=1
bernstein2_l, bernstein2_r	left and right truncation of Bernstein polynomial basis functions for rescaled frequency, 0<=bernstein2_l<bernstein2_r<=1

Value

A list of prior parameter values

beyondWhittle documentation built on May 31, 2023, 6:51 p.m.