genMixedData: Generate simulated mixed-type data with cluster structure.

In kamila: Methods for Clustering Mixed-Type Data

Description

This function simulates mixed-type data sets with a latent cluster structure, with continuous and nominal variables.

Usage

genMixedData(
  sampSize,
  nConVar,
  nCatVar,
  nCatLevels,
  nConWithErr,
  nCatWithErr,
  popProportions,
  conErrLev,
  catErrLev
)

Arguments

sampSize	Integer: Size of the simulated data set.
nConVar	The number of continuous variables.
nCatVar	The number of categorical variables.
nCatLevels	Integer: The number of categories per categorical variables. Currently must be a multiple of the number of populations specified in popProportions.
nConWithErr	Integer: The number of continuous variables with error.
nCatWithErr	Integer: The number of categorical variables with error.
popProportions	A vector of scalars that sums to one. The length gives the number of populations (clusters), with values denoting the prior probability of observing a member of the corresponding population. NOTE: currently only two populations are supported.
conErrLev	A scalar between 0.01 and 1 denoting the univariate overlap between clusters on the continuous variables specified to have error.
catErrLev	Univariate overlap level for the categorical variables with error.

Details

This function simulates mixed-type data sets with a latent cluster structure. Continuous variables follow a normal mixture model, and categorical variables follow a multinomial mixture model. Overlap of the continuous and categorical variables (i.e. how clear the cluster structure is) can be manipulated by the user. Overlap between two clusters is the area of the overlapping region defined by their densities (or, for categorical variables, the summed height of overlapping segments defined by their point masses). The default overlap level is 0.01 (i.e. almost perfect separation). A user-specified number of continuous and categorical variables can be specified to be "error variables" with arbitrary overlap within 0.01 and 1.00 (where 1.00 corresponds to complete overlap). NOTE: Currently, only two populations (clusters) are supported. While exact control of overlap between two clusters is straightforward, controlling the overlap between the K choose 2 pairwise combinations of clusters is a more difficult task.

Value

A list with the following elements:

trueID	Integer vector giving population (cluster) membership of each observation
trueMus	Mean parameters used for population (cluster) centers in the continuous variables
conVars	The continuous variables
errVariance	Variance parameter used for continuous error distribution
popProbsNoErr	Multinomial probability vectors for categorical variables without measurement error
popProbsWithErr	Multinomial probability vectors for categorical variables with measurement error
catVars	The categorical variables

Examples

dat <- genMixedData(100, 2, 2, nCatLevels=4, nConWithErr=1, nCatWithErr=1,
  popProportions=c(0.3,0.7), conErrLev=0.3, catErrLev=0.2)
with(dat,plot(conVars,col=trueID))
with(dat,table(data.frame(catVars[,1:2],trueID, stringsAsFactors = TRUE)))

Example output

, , trueID = 1

   X2
X1   1  2  3  4
  1 10  6  2  0
  2  6  6  0  0
  3  0  0  0  0
  4  0  0  0  0

, , trueID = 2

   X2
X1   1  2  3  4
  1  0  0  0  0
  2  0  0  0  0
  3  1  2 12 21
  4  3  3 14 14

kamila documentation built on March 13, 2020, 9:08 a.m.