TSDT (version 1.0.8)

subsample: subsample

Description

Generate a vector of subsamples.

Usage

subsample(
  x,
  trt = NULL,
  trt_control = "Control",
  training_fraction = NULL,
  validation_fraction = NULL,
  test_fraction = NULL,
  n_samples = 1
)

Value

Vector of objects of class Subsample.

Arguments

x

<Source data to subsample.

trt

Treatment variable. (optional)

trt_control

Value for treatment control arm. Defaulte value is 'Control'.

training_fraction

Fraction of source data to include in training subsample.

validation_fraction

Fraction of source data to include in validation subsample.

test_fraction

Fraction of source data to include in test subsample.

n_samples

Number of subsamples to generate.

Details

Each subsample will contain training, validation, and test data in proportions specified by the user. If a treatment variable is supplied the ratio of treatments will be preserved as closely as possible.

See Also

Subsample

Examples

## Generate example data frame containing response and treatment
N <- 50
x <- data.frame( runif( N ) )
names( x ) <- "response"
x$treatment <- factor( sample( c("Control","Experimental"), size = N,
                       prob = c(0.8,0.2), replace = TRUE ) )

## Generate two subsamples
ex1 <- subsample( x,
                  training_fraction = 0.9,
                  test_fraction = 0.1,
                  n_samples = 2 )

## Generate two subsamples preserving treatment ratio
ex2 <- subsample( x,
                  trt = x$treatment,
                  trt_control = "Control",
                  training_fraction = 0.7,
                  validation_fraction = 0.2,
                  test_fraction = 0.1,
                  n_samples = 2 )