mv.plots.SM: Diagnostic mean values plots.

In CLAST: Exact Confidence Limits after a Sequential Trial

Diagnostic mean values plots.

Description

Plots mean value of upper limit, lower limit and interval width for four different ranking methods. This function is basically a wrapper for mv.plot.

Usage

mv.plots.SM(n, a, b, type = "interval",
B = 100, offset = TRUE, plt = c(1, 1, 1), p0 = NULL, p1 = NULL, focus = FALSE)

Arguments

n	Design vector of planned sample sizes
a	Design vector of lower futility boundaries
b	Design vector of upper superiority boundaries
type	Either "upper", "lower" or "interval" (default)
B	Integer controlling fineness of plot (default=100)
offset	if TRUE then ML mean value is subtracted
plt	Logical vector indicating output plots of upper, lower and interval (default=c(1,1,1))
p0	Lower (null) benchmark for success probability
p1	Upper (alternative) benchmark for success probability
focus	Logical. If true, plots are restricted to p between p0 and p1. (default=FALSE)

Value

NULL

Author(s)

Chris J. Lloyd

References

Lloyd, C.J. (2021) Exact confidence limits after a group sequential single arm binary trial. Statistics in Medicine, Volume 38, 2389-2399. doi: 10.1002/sim.8909

Examples

# Figure 2 in Lloyd (2020)
  n=c(5,6,5,9)
  a=c(2,4,5,12)
  b=c(5,9,11,13)
  p0=.4
  p1=.75
  mv.plots.SM(n,a,b,p0=p0,p1=p1)
# Produces three panel graphic identical to Figure 2 in reference
  mv.plots.SM(n,a,b,p0=p0,p1=p1,focus=TRUE)
# Produces alternative graphic focussed on relevant values of p.
# In both cases LR (in blue) appears best. CP can perform poorly
# for values of p outside the range of interest.

CLAST documentation built on April 8, 2022, 9:06 a.m.