composite: Score Composition

In adoptr: Adaptive Optimal Two-Stage Designs in R

Description

composite defines new composite scores by point-wise evaluation of scores in any valid numerical expression.

Usage

composite(expr, label = NA_character_)

## S4 method for signature 'CompositeScore,TwoStageDesign'
evaluate(s, design, ...)

Arguments

expr	Expression (in curly brackets); must contain at least one score variable; if multiple scores are used, they must either all be conditional or unconditional. Currently, no non-score variables are supported
label	object label (string)
s	object of class CompositeScore
design	object
...	further optional arguments

Value

an object of class CompositeConditionalScore or CompositeUnconditionalScore depending on the class of the scores used in expr

See Also

Scores

Examples

ess   <- ExpectedSampleSize(Normal(), PointMassPrior(.4, 1))
power <- Power(Normal(), PointMassPrior(.4, 1))

# linear combination:
composite({ess - 50*power})

# control flow (e.g. for and while loops)
composite({
  res <- 0
  for (i in 1:3) {
     res <- res + ess
  }
  res
})

# functional composition
composite({log(ess)})
cp <- ConditionalPower(Normal(), PointMassPrior(.4, 1))
composite({3*cp})

adoptr documentation built on June 28, 2021, 5:11 p.m.