Time To Event Target Data — TimeToEventTargetData • BExTE

This class represents target data for time-to-event analysis. It inherits from the TargetData class.

The target trial follows the modified fixed-calendar design: patients enter uniformly over an accrual window, each may be followed for at most max_follow_up_time, and the database closes final_follow_up after the end of recruitment. See R/simulation_time_to_event.R for the generator and the Cox fit.

Super class

TargetData -> TimeToEventTargetData

Public fields

control_rate: Rate in the control arm of the target study
treatment_rate: Rate in the treatment arm of the target study
max_follow_up_time: Maximum individual follow-up time, L
accrual_period: Length of the recruitment window, A
final_follow_up: Time from the end of recruitment to database closure, F
dropout_probability: Probability of loss to follow-up over the maximum follow-up time
event_time_distribution: Either "exponential" or "weibull"
weibull_shape: Common Weibull shape parameter, q
weibull_scale: Control-arm Weibull scale, calibrated to the reported relapse-free probability
treatment_delay: Time before the treatment effect starts, in years; 0 for proportional hazards
late_log_hr: Log hazard ratio once the effect has started, chosen so that the Cox model's large-sample limit equals the treatment effect

Methods

Inherited methods

TargetData$plot_sample()

`TimeToEventTargetData$new()`

Initialize the TimeToEventTargetData object

Usage

TimeToEventTargetData$new(
  source_data,
  sampling_approximation,
  target_sample_size_per_arm,
  control_drift = 0,
  treatment_drift,
  summary_measure_likelihood,
  max_follow_up_time,
  accrual_period = NULL,
  final_follow_up = NULL,
  weibull_shape = NULL,
  weibull_relapse_free_probability = NULL,
  dropout_probability = 0,
  event_time_distribution = "exponential",
  treatment_delay = 0
)

Arguments

source_data: The source data used for generating the target data.
sampling_approximation: A flag indicating whether to use sampling approximation.
target_sample_size_per_arm: The target sample size per arm.
control_drift: The control drift.
treatment_drift: The treatment drift.
summary_measure_likelihood: The summary measure distribution.
max_follow_up_time: The maximum individual follow-up time.
accrual_period: The length of the recruitment window.
final_follow_up: The time between the end of recruitment and database closure.
weibull_shape: The common Weibull shape parameter.
weibull_relapse_free_probability: The control-arm relapse-free probability at the maximum follow-up time, used to calibrate the Weibull scale.
dropout_probability: The probability of loss to follow-up over the maximum follow-up time.
event_time_distribution: Either "exponential" or "weibull".
treatment_delay: Time before the treatment effect starts, in years. Zero, the default, is the proportional-hazards design.

`TimeToEventTargetData$dropout_rate()`

The loss-to-follow-up rate implied by the dropout probability.

Usage

TimeToEventTargetData$dropout_rate()

Returns

A single number, zero when there is no loss to follow-up.

`TimeToEventTargetData$arm_parameters()`

The event-time parameters of each arm: rates under the exponential model, scales under the Weibull model.

Usage

TimeToEventTargetData$arm_parameters()

Returns

A list with elements control and treatment. Generate target data

`TimeToEventTargetData$generate()`

This function generates target data based on the specified parameters.

Usage

TimeToEventTargetData$generate(n_replicates)

Arguments

n_replicates: The number of replicates to generate.

Returns

A data frame containing the generated target data.

`TimeToEventTargetData$to_dict()`

Converts the target data object to a dictionary.

Usage

TimeToEventTargetData$to_dict()

Returns

A list representing the target data object.

`TimeToEventTargetData$clone()`

The objects of this class are cloneable with this method.

Usage

TimeToEventTargetData$clone(deep = FALSE)

Arguments

deep: Whether to make a deep clone.