Description Usage Arguments Details Value References Examples
getEffBounds
is based on asymptotic, multivariate normal distribution (also called canonical distribution) of test statistics (Rosenblum et al., 2016). getEffBounds_Maurer_Bretz_2013
uses method from Maurer Bretz (2013).
Let H01, H02 and H0C respectively denote the null hypotheses that there is no treatment effect in subpopulation 1, subpopulation 2 and the combined population.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 | getEffBounds(p1, r1, r2, var_s1_trt, var_s1_con, var_s2_trt, var_s2_con,
time_limit = 90, num_stages, n_total, n_per_stage, FWER,
H01_eff_allocated = NULL, H02_eff_allocated = NULL,
H0C_eff_allocated = NULL, FWER_allocation_matrix = NULL,
delta_eff = NULL, H01_eff_total_allocated = NULL,
H02_eff_total_allocated = NULL, H0C_eff_total_allocated = NULL, abseps,
maxpts, errtol, ...)
getEffBounds_Maurer_Bretz_2013(p1, r1, r2, var_s1_trt, var_s1_con, var_s2_trt,
var_s2_con, time_limit, num_stages, n_total = NULL, n_per_stage, FWER,
H01_eff_allocated = NULL, H02_eff_allocated = NULL,
H0C_eff_allocated = NULL, FWER_allocation_matrix = NULL,
delta_eff = NULL, H01_eff_total_allocated = NULL,
H02_eff_total_allocated = NULL, H0C_eff_total_allocated = NULL, abseps,
maxpts, errtol, graph_edge_12, graph_edge_2C, graph_edge_C1, ...)
|
p1 |
proportion of population in subpopulation 1. |
r1 |
probability of being randomized to treatment in subpopulation 1 |
r2 |
probability of being randomized to treatment in subpopulation 2 |
var_s1_trt |
variance of the outcome under treatment in subpopluation 1. |
var_s1_con |
variance of the outcome under control in subpopluation 1. |
var_s2_trt |
variance of the outcome under treatment in subpopluation 2. |
var_s2_con |
variance of the outcome under control in subpopluation 2. |
time_limit |
time limit for calculations |
num_stages |
number of stages for the trial |
n_total |
the total, maximum number of patients to recruit by the end of the study. If entered, n_per_stage will be scaled to have this sum. |
n_per_stage |
a vector with length equal to |
FWER |
Familywise Type I error rate for the trial. |
H01_eff_allocated |
a vector of length |
H02_eff_allocated |
a vector of length |
H0C_eff_allocated |
a vector of length |
FWER_allocation_matrix |
a matrix telling the proportion of Type I error to allocation to each hypothesis at each stage. If entered, this will override |
delta_eff |
This determines the allocation of Type I error across stages if |
H01_eff_total_allocated |
rather than setting the error allocated to each stage, the user can instead set the total error allocated to each hypothesis. |
H02_eff_total_allocated |
see |
H0C_eff_total_allocated |
see |
abseps |
passed to pmvnorm in determining precision of calculations. |
maxpts |
passed to pmvnorm in determining precision of calculations. |
errtol |
determines precision of calculation of z-score boundary. |
... |
needed so that function ignores unused arguments when called by |
graph_edge_12 |
(Maurer, Bretz; 2013) The proportion of alpha to reallocate from H01 to H02 in the event that H01 is rejected |
graph_edge_2C |
(Maurer, Bretz; 2013) The proportion of alpha to reallocate from H02 to H0C in the event that H02 is rejected |
graph_edge_C1 |
(Maurer, Bretz; 2013) The proportion of alpha to reallocate from H0C to H01 in the event that H0C is rejected |
getEffBounds
strongly controls the familywise Type I error rate, based on the
generalized error-spending approach that allocates alpha (Type I error)
across stages and populations using the M_COV multiple testing procedure from Rosenblum et al. (2016).
The algorithm for efficacy boundary construction involves sequential computation
of the multivariate normal distribution using the package mvtnorm.
A list of efficacy boundaries for the z-statistics corresponding to each null hypothesis.
Maurer, W. and Bretz, F. (2013). Multiple testing in group sequential trials using graphical approaches. Statistics in Biopharmaceutical Research.
Rosenblum, M., Qian, T., Du, Y., and Qiu, H., Fisher, A. (2016). Multiple Testing Procedures for Adaptive Enrichment Designs: Combining Group Sequential and Reallocation Approaches. Biostatistics. 17(4), 650-662. https://goo.gl/c8GlcH
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 | ## Not run:
# Fully allocate the error for each stage
K <- 5
getEffBounds(p1 = 0.33,
r1 = 1/2,
r2 = 1/2,
var_s1_trt = 0.375*(1-0.375),
var_s1_con = 0.25*(1-0.25),
var_s2_trt = 0.325*(1-0.325),
var_s2_con = 0.2*(1-0.2),
num_stages = 5,
n_total = NULL,
n_per_stage = rep(200,K),
FWER = 0.025,
abseps = 0.000001,
errtol = .01,
maxpts = 10000,
H01_eff_allocated=rep(0.025/(3*K),K),
H02_eff_allocated=rep(0.025/(3*K),K),
H0C_eff_allocated=rep(0.025/(3*K),K)
)
# Boundaries for Maurer Bretz 2013
getEffBounds_Maurer_Bretz_2013(p1 = 0.33,
r1 = 1/2,
r2 = 1/2,
var_s1_trt = 0.375*(1-0.375),
var_s1_con = 0.25*(1-0.25),
var_s2_trt = 0.325*(1-0.325),
var_s2_con = 0.2*(1-0.2),
num_stages = 5,
n_total = NULL,
n_per_stage = rep(200,K),
FWER = 0.025,
abseps = 0.000001,
errtol = .01,
maxpts = 10000,
graph_edge_12=0.5,
graph_edge_2C=0.5,
graph_edge_C1=0.5,
time_limit = 100,
H01_eff_allocated=rep(0.025/(3*K),K),
H02_eff_allocated=rep(0.025/(3*K),K),
H0C_eff_allocated=rep(0.025/(3*K),K)
)
## End(Not run)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.