oc_stein: Compute operating characteristics using STEIN

Description

oc_stein() uses the STEIN design to compute operating charateristics of a user-specificed trial scenario. This design uses target toxicity and efficacy rates separately to form the cutoff intervals within a decision map.

Usage

oc_stein(
  ndose,
  target_t,
  lower_e,
  ncohort = 10,
  cohortsize = 3,
  startdose = 1,
  OBD = 0,
  psi1 = 0.2,
  psi2 = 0.6,
  psafe = 0.95,
  pfutility = 0.9,
  ntrial = 10000,
  utilitytype = 1,
  u1,
  u2,
  prob = NULL
)

Value

A list containing operating characteristics such as:

bd.sel: OBD selection percentage
od.sel: Favorable dose selection percentage
bd.pts: Average percentage of patients at the OBD
od.pts: Average percentage of patients at the favorable doses
earlystop: Percentage of early stopped trials
overdose: Overdose patients percentage
poorall: Poor allocation percentage
ov.sel: Overdose selection percentage

Arguments

ndose

Integer. Number of dose levels. (Required)

target_t

Numeric. Target toxicity probability. (Required)

lower_e

Numeric. Minimum acceptable efficacy probability. (Required)

ncohort

Integer. Number of cohorts. (Default is 10)

cohortsize

Integer. Size of a cohort. (Default is 3)

startdose

Integer. Starting dose level. (Default is 1)

OBD

Integer. True index of the Optimal Biological Dose (OBD) for the trial scenario. (Default is 0)

If set to 0: Random OBD will be selected.
Other: Treat this argument as the true OBD.

psi1

Numerical. Highest inefficacious efficacy probability.

psi2

Numerical. Lowest highly-promising efficacy probability.

psafe

Numeric. Early stopping cutoff for toxicity. (Default is 0.95)

pfutility

Numeric. Early stopping cutoff for efficacy. (Default is 0.95)

ntrial

Integer. Number of random trial replications. (Default is 10000)

utilitytype

Integer. Type of utility structure. (Default is 1)

If set to 1: Use preset weights (w11 = 0.6, w00 = 0.4)
If set to 2: Use (w11 = 1, w00 = 0)
Other: Use user-specified values from u1 and u2.

u1

Numeric. Utility parameter w_11. (0-100)

u2

Numeric. Utility parameter w_00. (0-100)

prob

Fixed probability vectors. If not specified, a random scenario is used by default. Use this parameter to provide fixed probability vectors as a list of the following named elements:

pE: Numeric vector of efficacy probabilities for each dose level.
pT: Numeric vector of toxicity probabilities for each dose level.
obd: Integer indicating the index of the true Optimal Biological Dose (OBD).
mtd: Integer indicating the index of the true Maximum Tolerated Dose (MTD).

For example:

prob <- list(
  pE = c(0.4, 0.5, 0.6, 0.6, 0.6),
  pT = c(0.1, 0.2, 0.3, 0.4, 0.4),
  obd = 3,
  mtd = 2
)

Examples

Run this code

oc_stein(
  ndose = 5,
  target_t = 0.3,
  lower_e = 0.4,
  ntrial = 10,
)

Run the code above in your browser using DataLab