smartsize: sample size calculation

In SMARTAR: Sequential Multiple Assignment Randomized Trial and Adaptive Randomization

Description

Return a message that contains the estimated strategy-specified means and their confidence interval, as well as the asymptotic variance-covariance matrix for these estimates.

Usage

smartsize(
  sim = NULL,
  delta = NULL,
  df = NULL,
  alpha = 0.05,
  beta = 0.2,
  global = TRUE,
  family = c("gaussian", "binomial")[1]
)

Arguments

sim	A numeric matrix containing the values of treatment sequence-specific parameters to generate the SMART data, including the values of stage-specific treatments, intermediate outcome and final primary outcome.
delta	The standardized effect size for sample size calculation.
df	The degrees of freedom for the chisquare test.
alpha	Type I error rate.
beta	Type II error rate.
global	If TRUE then power the SMART based on a global test, otherwise power the SMART based on a pairwise test. The default is TRUE
family	A character string to specify the type of final primary outcome. The default is family=“gaussian”, which refers to the continuous primary outcome. If family=”binomial” then the primary outcome will be treated as binary variable.

Value

Standardized effect size and total sample size for SMART

• delta: standardized effect size

• n: total sample size

References

Murphy, S. A. (2005). An experimental design for the development of adaptive treatment strategies. Statistics in Medicine. 24(10): 1455-1481.

Ogbagaber S. B., Karp, J., and Wahed A.S. (2015). Design of sequentially randomization trials for testing adaptive treatment strategies. Statistics in Medicine. DOI: 10.1002/sim.6747.

SMARTAR documentation built on July 31, 2020, 1:06 a.m.