generate.cluster.data: Function to Generate Data According to the Supcluster Model

In supcluster: Supervised Cluster Analysis

Function to Generate Data According to the Supcluster Model

Description

Generates cluster data according to the used for supervised clustering

Usage

generate.cluster.data(ratio,npats=80,clusts=c(12,8,12,12,6),
        sig=1,gamma=1,beta=c(-5,-2.5,0,2.5,5),outcomeModel=NULL)

Arguments

ratio

The ratio τ^2/σ^2 of the variance of the\ random effects to the error variance of the features

npats	Number of observations in the data set.
clusts	The cluster identity of the features
sig	The error variance of the features.
gamma	The error variance of the outcome.
beta	The value of the regression coefficients
outcomeModel	A function that returns a data frame with npats observations and rows that depend on the data object chosen. Two outcomeModel programs are provided, binaryOutcome and survivalOutcome, however users can write their own outcome model. If NULL no data object is returned

Value

A list with one element if outcomeModel=NULL which is a data frame which is npats times ngens+1 the last column is the outcome. Otherwise a list of two data frames, one being the feature data and the other being the outcome data according to what outcomeModel is used.

Author(s)

David A. Schoenfeld

See Also

supcluster

supcluster documentation built on May 20, 2022, 1:07 a.m.