dlt_rate: Evaluation of the DLT rate

In dnzmarcio/ewoc: Escalation with Overdose Control

Description

Calculate the DLT rate for each trial, the average DLT rate, the percent of trials which have DLT rate > target_rate + margin, the percent of trials which have DLT rate < target_rate - margin and the percent of trials which have target_rate - margin < DLT rate < target_rate + margin.

Usage

dlt_rate(
  dlt_matrix,
  trial = FALSE,
  target_rate = NULL,
  margin = NULL,
  digits = 2
)

Arguments

dlt_matrix	a matrix of the number of DLT for each step of the trial (column) and for each trial (row).
trial	a logical value indicating if the DLT rate for each trial should be returned.
target_rate	a numerical value of the target rate of DLT.
margin	a numerical value of the acceptable distance from the target_rate.
digits	a numerical value indicating the number of digits.

Value

trial a numerical vector of the DLT rate for each trial.

average a numerical value of the average of DLT rate considering a batch of trials.

upper the percent of trials which the DLT rate > target_rate + margin if margin != NULL and target_rate != NULL.

lower the percent of trials which the DLT rate < target_rate - margin if margin != NULL and target_rate != NULL.

interval the percent of trials which the target_rate - margin < DLT rate < target_rate + margin if margin != NULL and target_rate != NULL.

Examples

## Not run: 
DLT <- 0
dose <- 20
step_zero <- ewoc_d1classical(DLT ~ dose, type = 'discrete',
                           theta = 0.33, alpha = 0.25,
                           min_dose = 0, max_dose = 100,
                           dose_set = seq(0, 100, 20),
                           rho_prior = matrix(1, ncol = 2, nrow = 1),
                           mtd_prior = matrix(1, ncol = 2, nrow = 1),
                           rounding = "nearest")
stop_rule_sim(step_zero)
response_sim <- response_d1classical(rho = 0.05, mtd = 20, theta = 0.33,
                                  min_dose = 10, max_dose = 50)
sim <- ewoc_simulation(step_zero = step_zero,
                       n_sim = 1, sample_size = 2,
                       alpha_strategy = "increasing",
                       response_sim = response_sim,
                       stop_rule_sim = stop_rule_sim,
                       ncores = 2)
dlt_rate(sim$dlt_sim)

## End(Not run)

## Not run: 
DLT <- 0
dose <- 20
step_zero <- ewoc_d1classical(DLT ~ dose, type = 'discrete',
                           theta = 0.33, alpha = 0.25,
                           min_dose = 0, max_dose = 100,
                           dose_set = seq(0, 100, 20),
                           rho_prior = matrix(1, ncol = 2, nrow = 1),
                           mtd_prior = matrix(1, ncol = 2, nrow = 1),
                           rounding = "nearest")
stop_rule_sim(step_zero)
response_sim <- response_d1classical(rho = 0.05, mtd = 20, theta = 0.33,
                                  min_dose = 10, max_dose = 50)
sim <- ewoc_simulation(step_zero = step_zero,
                       n_sim = 2, sample_size = 30,
                       alpha_strategy = "increasing",
                       response_sim = response_sim,
                       stop_rule_sim = stop_rule_sim,
                       ncores = 2)
dlt_rate(sim$dlt_sim)

## End(Not run)

dnzmarcio/ewoc documentation built on Sept. 28, 2021, 6:52 a.m.