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 class '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
Examples
ess <- ExpectedSampleSize(Normal(), PointMassPrior(.4, 1))
power <- Power(Normal(), PointMassPrior(.4, 1))
# linear combination:
composite({ess - 50*power})
#> E[n(x1)] - 50 * Pr[x2>=c2(x1)]
# control flow (e.g. for and while loops)
composite({
res <- 0
for (i in 1:3) {
res <- res + ess
}
res
})
#> res <- 0; for (i in 1:3) {
#> res <- res + E[n(x1)]
#> }; res
# functional composition
composite({log(ess)})
#> log(E[n(x1)] )
cp <- ConditionalPower(Normal(), PointMassPrior(.4, 1))
composite({3*cp})
#> 3 * Pr[x2>=c2(x1)|x1]