EW_design_initial_GLM {ForLion} | R Documentation |
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",
delta0 = 1e-06,
epsilon = 1e-12,
maxit = 1000
)
Arguments
k.continuous |
number of continuous variables |
factor.level |
list of distinct factor levels, “(min, max)” for continuous factors that always come first, finite sets for discrete factors. |
Integral_based |
TRUE or FALSE, whether or not integral-based EW D-optimality is used, FALSE indicates sample-based EW D-optimality is used. |
b_matrix |
matrix of bootstrapped or simulated parameter values. |
joint_Func_b |
prior distribution function of model parameters |
Lowerbounds |
vector of lower ends of ranges of prior distribution for model parameters. |
Upperbounds |
vector of upper ends of ranges of prior distribution for model parameters. |
xlist_fix |
list of discrete factor experimental settings under consideration, default NULL indicating a list of all possible discrete factor experimental settings will be used. |
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" |
delta0 |
tuning parameter, the distance threshold, || x_i(0) - x_j(0) || >= delta0 |
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