EW_design_initial_GLM: function to generate a initial EW Design for generalized...

In ForLion: 'ForLion' Algorithms to Find Optimal Experimental Designs with Mixed Factors

function to generate a initial EW Design for generalized linear models

Description

function to generate a initial EW Design for generalized linear models

Usage

EW_design_initial_GLM(
  k.continuous,
  factor.level,
  Integral_based,
  b_matrix,
  joint_Func_b,
  Lowerbounds,
  Upperbounds,
  xlist_fix = NULL,
  lvec,
  uvec,
  h.func,
  link = "continuation",
  delta = 1e-06,
  epsilon = 1e-12,
  maxit = 1000
)

Arguments

k.continuous	number of continuous variables
factor.level	lower, upper limit of continuous variables, and discrete levels of categorical variables, continuous factors come first
Integral_based	TRUE or FALSE, if TRUE then we will find the integral-based EW D-optimality otherwise we will find the sample-based EW D-optimality
b_matrix	The matrix of the sampled parameter values of beta
joint_Func_b	The prior joint probability distribution of the parameters
Lowerbounds	The lower limit of the prior distribution for each parameter
Upperbounds	The upper limit of the prior distribution for each parameter
xlist_fix	the restricted discrete settings to be chosen, default to NULL, if NULL, will generate a discrete uniform random variables
lvec	lower limit of continuous variables
uvec	upper limit of continuous variables
h.func	function, is used to transfer the design point to model matrix (e.g. add interaction term, add intercept)
link	link function, default "continuation", other options "baseline", "adjacent" and "cumulative"
delta	tuning parameter, the distance threshold, || x_i(0) - x_j(0) || >= delta
epsilon	determining f.det > 0 numerically, f.det <= epsilon will be considered as f.det <= 0
maxit	maximum number of iterations

Value

X matrix of initial design point

p0 initial random approximate allocation

f.det the determinant of the expected Fisher information matrix for the initial design

ForLion documentation built on June 10, 2025, 5:13 p.m.