Methods for calculating estimates, confidence intervals, and p-values from binary sequential boundaries

Description

For routine use, these functions will not need to be called directly but are called from within the design functions (see designOBF). If needed, use analyze for any class representing a binary sequential boundary (see bound-class), and the appropriate function is called.

Usage

1
2
3
4
5
6
analyzeBound(object, theta0 = 0.5, stats = "all", 
    alternative = "two.sided", conf.level = 0.95, 
    tsalpha = NULL, ...)
analyzeBoundNBF(object, theta0 = 0.5, stats = "all", 
    alternative = "two.sided", conf.level = 0.95, 
    tsalpha = NULL, cipMatch = TRUE, ...)

Arguments

object

a binary sequential boundary (for classes see bound-class)

theta0

probability of success under the null

stats

character, either 'all' or 'pval'

tsalpha

vector of length 2, error on either side, if present overrides alternative and conf.level (see details)

alternative

character, either 'two.sided', 'less', or 'greater'

conf.level

confidence level

cipMatch

logical, for non-binding futility boundaries, should CI match the p-values on the binding boundary

...

further arguments to be passed

Value

if stats='all' returns an object of class 'boundEst', otherwise returns a numeric vector of p-values

See Also

See analyze-methods