ConditionalSampleSize-class: (Conditional) Sample Size of a Design

In kkmann/otsd: Adaptive Optimal Two-Stage Designs in R

(Conditional) Sample Size of a Design

Description

This score simply evaluates n(d, x1) for a design d and the first-stage outcome x1. The data distribution and prior are only relevant when it is integrated.

Usage

ConditionalSampleSize(label = "n(x1)")

ExpectedSampleSize(dist, prior, label = "E[n(x1)]")

## S4 method for signature 'ConditionalSampleSize,TwoStageDesign'
evaluate(s, design, x1, optimization = FALSE, ...)

Arguments

label	object label (string)
dist	a univariate distribution object
prior	a Prior object
s	Score object
design	object
x1	stage-one test statistic
optimization	logical, if TRUE uses a relaxation to real parameters of the underlying design; used for smooth optimization.
...	further optional arguments

See Also

Scores

Examples

design <- TwoStageDesign(50, .0, 2.0, 50, 2.0, order = 5L)
prior  <- PointMassPrior(.4, 1)

css   <- ConditionalSampleSize()
evaluate(css, design, c(0, .5, 3))

ess   <- ExpectedSampleSize(Normal(), prior)

# those two are equivalent
evaluate(ess, design)
evaluate(expected(css, Normal(), prior), design)

kkmann/otsd documentation built on Feb. 3, 2024, 4:50 p.m.