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.
smartsize(sim = NULL, delta = NULL, df = NULL, alpha = 0.05,
beta = 0.2, global = TRUE, family = "normal", method = "Gest")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.
The standardized effect size for sample size calculation.
The degrees of freedom for the chisquare test.
Type I error rate.
Type II error rate.
If TRUE then power the SMART based on a global test, otherwise power the SMART based on a pairwise test. The default is TRUE
A character string to specify the type of final primary outcome. The default is family=<U+201C>normal<U+201D>, which refers to the continuous primary outcome. If family=<U+201D>binomial<U+201D> then the primary outcome will be treated as binary variable.
A characteristic string to specify the method of estimation. We provide 2 method of estimation. The default is method=<U+201D>MLE<U+201D>, which is the Maximized Likelihood Estimation method. If method=<U+201D>IPWN<U+201D> then use the Normalized Inversed Probability Weighted method. The input of this option is only needed when power the SMART based on global test but not the pairwise test.
Standardized effect size and total sample size for SMART
delta: standardized effect size
n: total sample size
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.