Fusion: Validation of estimates of conditional average treatment...

In duolajiang/RCTrep: Validation of Estimates of Treatment Effects in Observational Data

Validation of estimates of conditional average treatment effects in objects of class TEstimator and SEstimator.

Description

Validation of estimates of conditional average treatment effects in objects of class TEstimator and SEstimator.

Validation of estimates of conditional average treatment effects in objects of class TEstimator and SEstimator.

Value

an R6 object

Methods

Public methods

• Fusion$new()

• Fusion$plot()

• Fusion$print()

• Fusion$evaluate()

• Fusion$clone()

---

Method new()

Usage

Fusion$new(..., stratification = NULL, stratification_joint = NULL)

Arguments

...

objects of class TEstimator and SEstimator.

stratification

a character vector specifying variables. The variables are used to select subgroups individually or in combination depending on stratification_joint. Default value is NULL.

stratification_joint

a logical indicating if subgroups are selected based on levels of individual variable in stratification or levels of combined variables in stratifiation. Default value is NULL.

---

Method plot()

Usage

Fusion$plot()

---

Method print()

Usage

Fusion$print()

---

Method evaluate()

Usage

Fusion$evaluate()

---

Method clone()

The objects of this class are cloneable with this method.

Usage

Fusion$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Examples

source.data <- RCTrep::source.data[sample(dim(RCTrep::source.data)[1],500),]
target.data <- RCTrep::target.data[sample(dim(RCTrep::target.data)[1],500),]

vars_name <- list(confounders_treatment_name = c("x1","x2","x3","x4","x5","x6"),
                  treatment_name = c('z'),
                  outcome_name = c('y')
)
confounders_sampling_name <- c("x2","x6")

source.obj <- TEstimator_wrapper(
 Estimator = "G_computation",
 data = source.data,
 vars_name = vars_name,
 outcome_method = "glm",
 outcome_form=y ~ x1 + x2 + x3 + z + z:x1 + z:x2 +z:x3+ z:x6,
 name = "RWD",
 data.public = FALSE
)

target.obj <- TEstimator_wrapper(
 Estimator = "Crude",
 data = target.data,
 vars_name = vars_name,
 name = "RCT",
 data.public = FALSE,
 isTrial = TRUE
)

strata <- c("x1","x4")
source.rep.obj <- SEstimator_wrapper(Estimator = "Exact",
                                     target.obj = target.obj,
                                     source.obj = source.obj,
                                     confounders_sampling_name =
                                     confounders_sampling_name)
source.rep.obj$EstimateRep(stratification = strata, stratification_joint = TRUE)

fusion <- Fusion$new(target.obj,
                     source.obj,
                     source.rep.obj)
fusion$plot()
fusion$evaluate()

duolajiang/RCTrep documentation built on April 15, 2023, 2:56 a.m.