gaussian_mixture_coord_exch: Coordinate exchange algorithm for a mixture model assuming...

Description Usage Arguments Value

View source: R/gaussian_model_functions.R

Description

Performs the coordinate exchange algorithm for a Scheffé mixture model. It can have many different random starts, or the coordinate exchange algorithm can be performed in a user-supplied matrix.

Usage

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
gaussian_mixture_coord_exch(
  n_runs = NULL,
  q = NULL,
  n_random_starts = 100,
  X = NULL,
  order = 1,
  opt_method = "B",
  max_it = 10,
  tol = 1e-04,
  n_cox_points = NULL,
  plot_designs = F,
  verbose = 1,
  opt_crit = 0,
  seed = NULL,
  n_cores = 1
)

Arguments

n_runs

number of runs

q

number of ingredient proportions

n_random_starts

number or random starts. Defaults to 100.

X

User supplied design matrix. Must be of size (n_runs, q) where: n_runs is the number of runs q is the number of ingredient proportions.

order

Order of the Scheffé model (1, 2, or 3).

opt_method

Optimization method in each step of the coordinate exchange algorithm. It can be "B" (Brent's algorithm) or "D" (discretization of Cox direction)

max_it

integer for maximum number of iterations that the coordinate exchange algorithm will do

tol

A positive error tolerance in Brent's method.

n_cox_points

number of points to use in the discretization of Cox direction

plot_designs

boolean. If TRUE, shows a plot of the initial and the final design. Only works if q is 3.

verbose

level of verbosity.

opt_crit

optimality criterion: D-optimality ("D" or 0) or I-optimality ("I" or 1)

seed

Seed for reproducibility

n_cores

Number of cores for parallel processing

Value

List with the following elements:


mariobecerra/opdesmixr documentation built on Aug. 13, 2021, 9:44 p.m.