randomizeR (version 3.0.1)

evaluate: Evaluation of several randomization procedures with respect to certain desirability functions applied to specified issues.

Description

Evaluation of several randomization procedures with respect to certain desirability functions applied to specified issues.

Usage

evaluate(..., statistic)
# S4 method for missing
evaluate(..., statistic)
# S4 method for character
evaluate(..., statistic)

Value

S4 object of class evaluation Comparison of randomization procedures with respect to desirability functions applied to specified issues, summarized by a prespecified statistic.

Arguments

...

at least one object of the class desScores or a list of objects of the class desScores.

statistic

character string that specifies on the basis of which statistic the evaluate function should be applied. The statistic can be chosen from "mean", "median", "min" or "max".

Details

The evaluate function allows the user to compare and evaluate different randomization procedures. It expects a number of objects that result when applying the getDesScores function to an assess object and specified desirability functions. The evaluate function summarizes the desirability scores of each randomization procedure on the basis of a prespecified statistic and incorporates them into a data frame. If no statistic is specified then it is automatically set to mean. If the function is applied to only one object it corresponds simply to summary(getDesScores(...)).

References

D. Schindler Assessment of Randomization Procedures in the Presence of Selection and Chronological Bias. PhD Thesis.

See Also

Representation of randomization procedures: randPar

Generation of randomization sequences: genSeq

issues for the desirability of randomization sequences

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

Examples

# compare Random Allocation Rule to Big Stick Design with respect to different issues
# and their corresponding desirability functions
issue1 <- corGuess("CS")
issue2 <- corGuess("DS")
RAR <- getAllSeq(rarPar(4))
BSD <- getAllSeq(bsdPar(4, mti = 2))
A1 <- assess(RAR, issue1, issue2)
A2 <- assess(BSD, issue1, issue2)

d1 <- derFunc(TV = 0.1, 0.7, 2)
d2 <- derFunc(0.5, c(0.3, 0.8), c(1, 1))
DesScore <- getDesScores(A1, d1, d2, weights = c(5/6, 1/6))
DesScore2 <- getDesScores(A2, d1, d2, weights = c(5/6, 1/6))

evaluate(DesScore, DesScore2)
evaluate(DesScore, DesScore2, statistic = "max")