set_switch_weight_model function

Set switching weight model

Description

set_switch_weight_model(object, numerator, denominator, model_fitter, ...)

## S4 method for signature 'trial_sequence'
set_switch_weight_model(
  object,
  numerator,
  denominator,
  model_fitter,
  eligible_wts_0 = NULL,
  eligible_wts_1 = NULL
)

## S4 method for signature 'trial_sequence_ITT'
set_switch_weight_model(object, numerator, denominator, model_fitter)

Arguments

• object: A trial_sequence object.
• numerator: Right hand side formula for the numerator model
• denominator: Right hand side formula for the denominator model
• model_fitter: A te_model_fitter object, such as stats_glm_logit
• ...: Other arguments used by methods.
• eligible_wts_0: Name of column containing indicator (0/1) for observation to be excluded/included in weight model.
• eligible_wts_1: Exclude some observations when fitting the models for the inverse probability of treatment weights. For example, if it is assumed that an individual will stay on treatment for at least 2 visits, the first 2 visits after treatment initiation by definition have a probability of staying on the treatment of 1.0 and should thus be excluded from the weight models for those who are on treatment at the immediately previous visit. Users can define a variable that indicates that these 2 observations are ineligible for the weight model for those who are on treatment at the immediately previous visit and add the variable name in the argument eligible_wts_1. Similar definitions are applied to eligible_wts_0 for excluding observations when fitting the models for the inverse probability of treatment weights for those who are not on treatment at the immediately previous visit.

Returns

object is returned with @switch_weights set

Examples

trial_sequence("PP") |>
  set_data(data = data_censored) |>
  set_switch_weight_model(
    numerator = ~ age_s + x1 + x3,
    denominator = ~ x3 + x4,
    model_fitter = stats_glm_logit(tempdir())
  )