IWFMult: Cocktail Algorithm implementation for I-Optimality and...

In optedr: Calculating Optimal and D-Augmented Designs

Cocktail Algorithm implementation for I-Optimality and A-Optimality (with matB = diag(k))

Description

Function that calculates the I-Optimal designs given the matrix B (should be integral of the information matrix over the interest region), or A-Optimal if given diag(k). The rest of the parameters can help the convergence of the algorithm.

Usage

IWFMult(
  init_design,
  grad,
  matB,
  min,
  max,
  grid.length,
  join_thresh,
  delete_thresh,
  delta_weights,
  tol,
  tol2
)

Arguments

init_design	optional dataframe with the initial design for the algorithm. A dataframe with two columns: Point contains the support points of the design. Weight contains the corresponding weights of the Points.
grad	function of partial derivatives of the model.
matB	optional matrix of dimensions k x k, integral of the information matrix of the model over the interest region for I-optimality.
min	numeric value with the inferior bound of the space of the design.
max	numeric value with the upper bound of the space of the design.
grid.length	numeric value that gives the grid to evaluate the sensitivity function when looking for a maximum.
join_thresh	numeric value that states how close, in real units, two points must be in order to be joined together by the join heuristic.
delete_thresh	numeric value with the minimum weight, over 1 total, that a point needs to have in order to not be deleted from the design.
delta_weights	numeric value in (0, 1), parameter of the algorithm.
tol	numeric value for the convergence of the weight optimizing algorithm.
tol2	numeric value for the stop condition of the algorithm.

Value

list correspondent to the output of the correspondent algorithm called, dependent on the criterion. A list of two objects:

• optdes: a dataframe with the optimal design in two columns, Point and Weight.

• sens: a plot with the sensitivity function to check for optimality of the design.

See Also

Other cocktail algorithms: DWFMult(), DsWFMult(), WFMult()

optedr documentation built on Nov. 18, 2022, 5:12 p.m.