| CreateData {RTaxometrics} | R Documentation |
Creates a data set
Description
This function creates an artificial data set based on either dimensional or categorical latent structure, which can vary according to a number of basic parameters. Such data can be useful for getting to know the taxometric programs and becoming familiar with their output by conducting analyses using data sets whose parameters are known.
Usage
CreateData(str, n = 600, k = 4, p = 0.5, d = 2, r = 0, r.tax = 0, r.comp = 0,
g = 0, h = 0, cuts = 0, uniform = F, seed = 1)
Arguments
str |
The type of data to be generated. Specify either "dim" for dimensional data or "cat" (or anything else) for categorical data. |
n |
Sample size. The default value is 600. |
k |
Number of variables. The default value is 4. |
p |
Taxon base rate. The default value is .5. |
d |
Standardized mean difference between groups. The default value is 2. |
r |
Correlation among variables. The default value is 0. |
r.tax |
Correlation among variables within the taxon. The default value is 0. |
r.comp |
Correlation among variables within the complement. The default value is 0. |
g |
Parameter used to control asymmetry (scalar); sign indicates direction and absolute value indicates magnitude of skew (e.g., +/- .30 yields substantial asymmetry). |
h |
Parameter used to control tail weight (scalar); positive values yield tails that are longer/thinner than a standard normal curve, negative values do the reverse (e.g., +/- .15 is a substantial departure from normality). |
cuts |
Parameter used to create ordered categorieas, if nonzero (scalar); number of categories will be cuts + 1. |
uniform |
Whether to generate random values (the program default) or use uniformly distributed quantiles (T/F). |
seed |
Random number seed; specifying the same seed enables users to generate and analyze identical data sets. The default value is 1. |
Details
Users should call this function directly if they wish to create an artificial data set.
Value
Data matrix; k columns contain data, final column contains classification.
Author(s)
John Ruscio <ruscio@tcnj.edu> and Shirley Wang <shirleywang@g.harvard.edu> Maintainer: John Ruscio <ruscio@tcnj.edu>
Examples
# creates a categorical data set
test.cat <- CreateData("cat")
# creates a dimensional data set
test.dim <- CreateData("dim")