derFunc: Representing Derringer-Suich desirability functions

In randomizeR: Randomization for Clinical Trials

Representing Derringer-Suich desirability functions

Description

Represents the Derringer-Suich desirability approach.

Usage

derFunc(TV, SLs, b)

Arguments

TV	numeric specifying the optimal desired value called the target value.
SLs	numeric vector of length at most 2 specifying the lower and/or upper specified border.
b	numeric vector of length at most 2 specifying the weight(s) for the punishment of deviations from the target value.

Details

derFunc represents the framework for left, right and two-sided desirability functions introduced by Derringer and Suich (1980). For all three different kinds of desirability functions the parameter TV must be specified. If the parameter SLs has length 1, either the left- or right-sided desirability function is created depending from whether the value is smaller (left-sided) or greater (right-sided) than the target value. By specifying SLs as a vector of length 2 a two-sided desirability function is created where the lower specified border is determined as the smaller value of SLs and thus the upper specified border is determined as the greater value. If there are no values specified for the weights, then they are automatically set to 1 (linear loss).

Value

S4 object of class derFunc, a formal representation of desirability functions introduced by Derringer and Suich.

References

Derringer, G., and Suich, R., (1980) Simultaneous Optimization of Several Response Variables. Journal of Quality Technology, 12, 214-219.

See Also

Other desirability topics: evaluate(), getDesScores(), plotDes(), plotEv(), probUnDes()

Examples

# create an object of a left-sided desirability function
dLeft <- derFunc(0.5, 0.3, 2)

# create an object of a right-sided desirability function
dRight <- derFunc(0.5, 0.8, 1)

# create an object of a two-sided desirability function
dLR <- derFunc(0.5, c(0.3, 0.9), c(3, 1))

randomizeR documentation built on Sept. 19, 2023, 1:08 a.m.