build_tree: Tree building function for NUTS

In yanbowisc/SIMP: Bayesian Simultaneous Partial Envelope Model

Tree building function for NUTS

Description

Tree building function for NUTS

Usage

build_tree(
  A,
  j,
  aa,
  u,
  v,
  jj,
  eps,
  A00,
  aa0,
  lpd_func,
  grad_lpd_func,
  M_diag,
  Delta_max = 1000,
  ...
)

Arguments

A	Parameter A or B.
j	The index for which column of A or B to be updated.
aa	The momentum vector of the j-th column of A or B.
u	a random number used in NUTS.
v	a random number from discrete Unif-1, +1 used in NUTS.
jj	The index for the layer of the tree.
eps	Stepsize.
A00	A or B from the previous iteration.
aa0	The momentum vector from the previous iteration.
lpd_func	The log full conditional posterior density of A or B.
grad_lpd_func	The function to calculate the gradients of the log full conditional posterior density of A or B.
M_diag	Covariance matrix for our matrix parameter A or B.
Delta_max	The Delta_max parameter used in the Algorithm 6 of NUTS.
...	Other parameters that may be useful.

yanbowisc/SIMP documentation built on Oct. 30, 2022, 1:33 a.m.