Methods to access the stage-one, stage-two, or overall sample size of a TwoStageDesign. n1 returns the first-stage sample size of a design, n2 the stage-two sample size conditional on the stage-one test statistic and n the overall sample size n1 + n2. When dealing with time-to-event endpoints, the functions n1, n2 and n return the number of events instead of the sample size. Internally, objects of the class TwoStageDesign allow non-natural, real sample sizes to allow smooth optimization (cf. minimize for details). The optional argument round allows to switch between the internal real representation and a rounded version (rounding to the next positive integer).

n1(d, ...)

# S4 method for TwoStageDesign
n1(d, round = TRUE, ...)

n2(d, x1, ...)

# S4 method for TwoStageDesign,numeric
n2(d, x1, round = TRUE, ...)

n(d, x1, ...)

# S4 method for TwoStageDesign,numeric
n(d, x1, round = TRUE, ...)

# S4 method for OneStageDesign,numeric
n2(d, x1, ...)

# S4 method for GroupSequentialDesign,numeric
n2(d, x1, round = TRUE, ...)

Arguments

d

design

...

further optional arguments

round

logical should sample sizes be rounded to next integer?

x1

stage-one test statistic

See also

TwoStageDesign, see c2 for accessing the critical values

Examples

design <- TwoStageDesign(
   n1    = 25,
   c1f   = 0,
   c1e   = 2.5,
   n2    = 50,
   c2    = 1.96,
   order = 7L
)

n1(design) # 25
#> [1] 25
design@n1 # 25
#> [1] 25

n(design, x1 = 2.2) # 75
#> [1] 75