TSDT (version 1.0.7)

distribution: distribution

Description

Returns the distribution of values used to compute TSDT summary statistics.

Usage

distribution(object, statistic, subgroup = NULL, subsub = NULL)

Value

A vector containing the observed values for the specified subgroup

Arguments

object

An object of class TSDT

statistic

The desired statistic distribution

subgroup

The desired subgroup

subsub

A subset of the subgroup

Details

This function returns the distribution of all values used to compute summary statistics for superior subgroups identified by the TSDT algorithm. The summary statistics returned for a TSDT object include the mean subgroup size, mean response value, and median value of the scoring function. These statistics reported seperately for in-bag and out-of-bag data sets, and also stratified by treatment arm. This function can also provide the distribution of all cutpoints for a numeric splitting variable in a subgroup definition.

See Also

TSDT, summary-methods

Examples

set.seed(0)
N <- 200
continuous_response = runif( min = 0, max = 20, n = N )
trt <- sample( c('Control','Experimental'), size = N, prob = c(0.4,0.6),
               replace = TRUE )
X1 <- runif( N, min = 0, max = 1 )
X2 <- runif( N, min = 0, max = 1 )
X3 <- sample( c(0,1), size = N, prob = c(0.2,0.8), replace = TRUE )
X4 <- sample( c('A','B','C'), size = N, prob = c(0.6,0.3,0.1), replace = TRUE )
covariates <- data.frame( X1 )
covariates$X2 <- X2
covariates$X3 <- factor( X3 )
covariates$X4 <- factor( X4 )

## Create a TSDT object
ex1 <- TSDT( response = continuous_response,
            trt = trt, trt_control = 'Control',
            covariates = covariates[,1:4],
            inbag_score_margin = 0,
            desirable_response = "increasing",
            oob_score_margin = 0,
            min_subgroup_n_control = 5,
            min_subgroup_n_trt = 5,
            n_sample = 5 )

## Show summary statistics
summary( ex1 )

## Get the number of subjects in each superior in-bag subgroup
distribution( ex1, statistic = 'Inbag_Subgroup_Size' )

## Get the vector of subgroup sample sizes for a particular subgroup
distribution( ex1, statistic = 'Inbag_Subgroup_Size',
              subgroup = 'X1=xxxxx' )

## Get the observed cutpoints for the numeric splitting variables in a subgroup
distribution( ex1, statistic = 'Cutpoints', subgroup = 'X1=xxxxx' )

## If the subgroup definition has more than one numeric splitting variable you
## can retrieve the numeric cutpoints for the splitting variables individually
distribution( ex1, statistic = 'Cutpoints', subgroup = 'X1=xxxxx',
              subsub = 'X1