EW_dprime_func_self: Function to calculate dEu/dx in the gradient of d(x, Xi),...

In ForLion: 'ForLion' Algorithm to Find D-Optimal Designs for Experiments

Function to calculate dEu/dx in the gradient of d(x, Xi), will be used in EW_ForLion_MLM_func() function

Description

Function to calculate dEu/dx in the gradient of d(x, Xi), will be used in EW_ForLion_MLM_func() function

Usage

EW_dprime_func_self(
  xi,
  bvec_matrix,
  h.func,
  h.prime,
  inv.F.mat,
  EUx,
  link = "continuation",
  k.continuous
)

Arguments

xi	a vector of design point
bvec_matrix	the matrix of the bootstrap parameter values of beta
h.func	function, is used to transfer xi to model matrix (e.g. add interaction term, add intercept)
h.prime	function, is used to find dX/dx
inv.F.mat	inverse of F_Xi matrix, inverse of the Expectation of fisher information of current design w/o new point
EUx	EU_x matrix in the algorithm, get from EW_Fi_MLM_func() function
link	link multinomial link function, default is"continuation", other choices "baseline", "cumulative", and "adjacent"
k.continuous	number of continuous factors

Value

dEU/dx in the gradient of sensitivity function d(x, Xi)

ForLion documentation built on April 11, 2025, 5:38 p.m.